|DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/4/13 8:35 PM|
I have read the FAQs and checked for similar issues: Yes
My site's URL (web address) is: www.daniweb.com
Secret Phrase: I have studied the Help Center, read the FAQs and searched for similar questions.
A bit of history:
For a treat, please see the attached Excel spreadsheet.
What's happened since I last reached out to the Google Webmaster Central community:
Why I'm Posting Now:
|DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/4/13 8:46 PM|
As was mentioned for hitchhiker , do you noIndex low quality / thin pages to keep all your indexed pages rich in content?
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/4/13 9:00 PM|
No offense, but did you read everything she wrote?:
This site, or the one from the other thread is not the typical Panda type website that has been hit. These are large UGC type websites that have implemented many quality controls.I'm still meta noindexing threads with 0 replies, etc. etc.
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/4/13 9:43 PM|
I did miss that - I don't think that's far enough however
even pages that have 100's of words can be low quality thin content - keep in mind part of the algorithm tweak was to nock out scrapped content , where pages of junk text were used to fake the ranking.
If its a series of posts agreeing with the OP or quiet repetitious , that can easily come across as thin - non value adding content
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/4/13 9:47 PM|
Doesn't about dot com have a thing where it only indexes the post /question when it is marked as answers .
And then only indexes the question and the post marked as the answer?
You have to click through to other replies not marked as the answer.
I think that's how it works - like a self regulating system
|DaniWeb hit in November '12 and haven't recovered since||terry noname||6/4/13 9:52 PM|
Hi Dani, congratulations on the 1 million Plus members (WOW). And sorry about the extra pounds :(
I had a look at your site from a user point of view and found some issues.
Firstly the pop up (at the top of the page) don't load on my Google Chrome it says my browsers to slow and its my fault!! (Very unusual, I thought I'd point that out) the page rendered in 0.1661 seconds using 3.21 MB.
The second issue (from a new user point of view) was "start a new community discussion" that only displayed the login drop down after I had edited the title and chose the forum and started to edit the editor.
The title "pop ups" on your navigation are useful, but don't you think you loosing out on valuable optimization.
Text on the page carries more weight than in a link title (my opinion).
That's all for now, Good luck and I hope you loose the additional, unwanted weight soon ;)
|ets||6/4/13 11:52 PM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/5/13 2:20 AM|
Panda is really tricky problem to solve. I've been hit on two different sites, made major changes and still no recovery. It feels to me that once you have been hit by Panda it's very unlikely that you will ever recover, which seems unfair.
Here are a few of my views suggestions:
1) Linking to thin content from the homepage - the first link I saw on your homepage was to this page: http://www.daniweb.com/software-development/vbnet/threads/456006/excel-reader - which feels very thin / low quality. Can you link to high quality content from your homepage?
2) At the bottom of that page there is a tag section, there are about 70 tags (way too many) and most seem point less, e.g. if, how, to, etc, etc
Matt cutts has said he doesnt like tags, its creates a lot of duplicate pages, so I would delete them all, it feels radical but to break out of panda you need to do radical stuff.
3) I hear that you noindexing thin threads. I would go a step further and 404 thin pages. Maybe do a review and if 99% of threads don't get a reply in the first 30 days, simply delete them automatically after 30 days. Or something like that. Matt has said that you should 404 low quality page, noindex is not as good
4) You got 422,000 pages indexed in google, are they all high quality? If not can you delete the low quality ones?
I hope that helps.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/5/13 2:31 AM|
I think there's a point at which a UGC site owner can say they've fully exhausted all the realistic options (for the current UGC context) in terms of cleanup. Looking at Dani's site, it seems she's hit that mark - and probably surpassed it. She could tweak this ad infinitum, we all could - at what point does this become just an 'MRI for healthy people'.
Sadly (if my understanding is correct), in the process she's lost her entire team - she's now working 10x harder to do future management by herself. My feeling is the internet would be much poorer without her influence. It's sad that I'm doing tech queries all day, and rarely get her site anymore; I found it very useful previously.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/5/13 2:54 AM|
I might be wrong but it looks like you have noindexed all/a lot of your pages.
e.g. http://www.daniweb.com/hardware-and-software/1 etc, etc.
I'm using this tool https://chrome.google.com/webstore/detail/web-developer/bfbameneiokkgbdmiekhjnmfkcnldhhm?hl=en
Once installed go to: http://www.daniweb.com/hardware-and-software/ - all pages are showing index (but they are in the google index, so I dont know what is going on)
then use the extension: go to information - view meta tag information
It says: robots: noindex
The weird thing is that I cant see that in your source code, or on your robots.txt file. So you will need to get a more technical seo to check it out.
Hope it helps,
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/5/13 3:18 AM|
If she doesn't have it in source (she doesn't) and it's not in the robots.txt (it isn't) then most likely your extension is mis-reporting or she's applied 'no-index' in a rel attribute on a link to it.
Show's no evidence of no-index either.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/5/13 3:19 AM|
I should add that I've also de-indexed most of those pages, as part of my 'cleanup' for the purposes of trying to avoid the strange 21.5 update. So it may well be intentional.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/5/13 3:34 AM|
Watch this old video from matt on tags. http://www.youtube.com/watch?v=bYPX_ZmhLqg
I would delete all your tag pages. I know you have noindexed them all, but you are still linking to them a LOT. I dont think they are useful, they feel spammy, you cant get traffic to them from google, and matt recommends 404ing low quality pages. http://www.seroundtable.com/cutts-google-india-seo-15407.html 12.42 for a couple of mins.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/5/13 3:35 AM|
it is weird. I've checked a lot of sites and what chrome extension is correct on all the other pages, so I would recommend some more digging on this noindex issue.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/5/13 4:06 AM|
I was going to point you to John Mueller's post over in hitchhiker's thread but I see you've already been there. :-)
Really sorry to hear that you're on your own now.
I want to offer a different line of thinking -- don't know if I'll get slammed for it or what, but the idea being that sometimes it's not necessarily something wrong with the site but that Google is no longer giving the site a boost. Or maybe even Google has taken away some kind of boost that UGC sites were previously receiving. A kind of "look at this,we're giving these sites 10 points for all their content, but it's UGC -- let's change that 10 points to 5 points" -- and then rankings drop.
Meaning: if you feel you're doing everything right, maybe you're not doing anything wrong now and the site is ranking where it's ranking --- and you need to focus on what you're doing right and continue to create a great site.
Of course, I could be wrong. But I just wanted to mention a different line of thinking. My own line of thinking has moved towards that way -- not in a defeatist way, but moreso, "if traffic comes back, it comes back but I'm going to work on other ways to get traffic and improve the site." (There's actually a lot I can do to improve my site right now, but that's a whole other story! :-))
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/5/13 4:10 AM|
Have you tried the Panguin tool? It shows your Analytics against known algorithm updates: http://www.barracuda-digital.co.uk/panguin-tool/ Maybe it'll give you some new insights. But, yup, the problem is that Google doesn't announce all updates.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/5/13 4:24 AM|
One more thought: the sites that are now appearing instead of daniweb for your previous top keywords, are they still answering the searcher's query? Because if they are, then Google's going to be happy with that and there *might* not be much you can do to change that. In my own experience, where I dropped, the searcher's question is still being answered. Sure, there's one particular site in the top rankings where I feel my site could provide a better user experience. But the searcher's question is answered. Not much tweaking I can do there, except to keep moving forward with my site.
I think what I''m trying to say is "Don't give up because there may be a solution, but don't drive yourself crazy with it either."
|ets||6/5/13 5:02 AM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/5/13 5:05 AM|
Hey Suzanneh, (I bet I'm your favourite poster right now! /s)
I don't think anyone reasonable could argue with your logic: Not only is it practical for the recovery, it's practical for most long term goals!
Infact it just helped reminded me I've been on this forum a bit too long now, time to get back to work! :)
This was the issue - It seems (from their comments) that most people agree with you, you're not 'in a different line of thinking' for the P21.5 casualties. They seem to want to apply changes, they seem to understand year after year we need to improve. I just get the feeling that people in Dani's situation might be a little tired of hearing about how they need to focus more on the SEO aspect, and less on their community (which is being suggested here to some degree) Again, just my feeling, I obviously can't speak for anyone else in that regard.
We're really just asking for some simple intel/guidance that is specific to our situation, not just an overall 'seo standpoint' - which most of us are fully aware of.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/5/13 5:15 AM|
the sites that are now appearing instead of daniweb for your previous top keywords, are they still answering the searcher's query? Because if they are, then Google's going to be happy with that and there *might* not be much you can do to change that.
Perhaps some other pertinent questions would be:
Note: Seo in the 'simple' form is completely ok, putting your best foot forward etc. I'm talking about having to be 'really on top of the game', as we would have to be to survive at our audience level.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/5/13 9:29 AM|
While I can understand the idea of removing thin pages that add absolutely no value, it's not practical to completely remove all of our navigation pages just because they aren't the best pages for Google to send traffic to. We noindex our navigation pages: that should be enough. People still need a way of navigating around the site, and the site is meant to be built for the users and not for google. Just because Google doesn't like to send traffic to navigational pages doesn't meant that navigational pages shouldn't exist on the web.
I'm not necessarily looking for SEO suggestions here. I basically feel like I heard every possible suggestion in the world, and I implemented everything I felt was right. I feel like I did the best job that I could do working on putting our best foot forward for google. But being an SEO isn't my fulltime job.
Over the past few months, we've hit a million members, we've launched an API, we've made countless design and functionality tweaks, the moderation team has gone back and fixed over 80,000 malformed old posts from people who didn't know how to use BBCode correctly with our old system, we've implemented an endorsements system that has since become quite popular, and countless, countless other things.
IMHO, *that's* my job as community administrator. It isn't a requirement of my job to have to continuously evaluate which keywords we rank for, evaluate seo tools, and compare our SEO trends against algorithm updates. I couldn't even list off our top ten keywords if I had to, so I have no clue if we still rank for what we used to but just on a smaller scale, or if we rank for an entirely different set of keywords, etc. Who knows and who cares. :)
If you do a search for "daniweb" within Google Webmaster Central here, you'll find countless, countless posts from Level 1 up through Level 12 users saying that they have previously used and enjoyed DaniWeb, found us useful and helpful, complemented us on having a great site, and said that we were unfairly hit by Google. Ok, so maybe we were unfairly hit, but now what? :) IMHO, I can only do what I can do, and it's up to Google to figure out how to rank us. Maybe, just maybe, Google did get it wrong this time around because there are so many people in these forums alone who have been advocating on our behalf. It's google's job to show the results that people want to find, and people here are speaking out that they are upset they are seeing less of us than they'd like.
The reason I've taken so long to speak up since being hit back in November is because I really don't think that I can do anything else SEO wise on my end. I'm happy with how SEO'd my site is and I am currently focusing on other facets to the site that don't involve cuddly google animals. At this point I'm putting everything in the hands of Google to take what we are presenting to them and hopefully do right by us and all of our many supporters.
What else can I say?
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/5/13 1:22 PM|
If you do a search for "daniweb" within Google Webmaster Central here, you'll find countless, countless posts from Level 1 up through Level 12 users saying that they have previously used and enjoyed DaniWeb, found us useful and helpful, complemented us on having a great site, and said that we were unfairly hit by Google. Ok, so maybe we were unfairly hit, but now what? :) IMHO, I can only do what I can do, and it's up to Google to figure out how to rank us. Maybe, just maybe, Google did get it wrong this time around because there are so many people in these forums alone who have been advocating on our behalf. It's google's job to show the results that people want to find, and people here are speaking out that they are upset they are seeing less of us than they'd like.
You'lls also found countless people that have offered you tips - most of which you disregarded. Do whatever you want Dani - but you can't claim that you haven't had loads of free help here.
It doesn't matter if you have a million people sign a petition. Just take the resources that we all have available to us and make a great site. You're not special. It's an algorithm. Work with it - not against it.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/5/13 1:24 PM|
LOVE that the first thread that I hit (which I had to scroll down to see) is just someone spamming your forum
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/5/13 1:32 PM|
Ashley, looks like that spam was posted 15 minutes ago and if you look at her source it is set to noindex. In my opinion she is taking the correct action there to indicate to Google to not index that page. On a UGC website, it is impossible to remove all spam 100% of the time immediately after it was posted. I bet I could spam on this Google Groups forum and it would take hours before it gets removed (not that I would), and not that it would say this forum is of poor quality. Now your point might be a little different if that was from two weeks ago, and it happens all over here website, but it happened 15 minutes ago, and very well could have been you, who knows!
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/5/13 1:37 PM|
Sure - spam/UGC is one thing. But Dani has lots of opportunities to improve her site and has had countless hours/help given to her. If she chooses to implement, fine. But you can't go through and said you did 1 thing - ignored 20 - and ask for a special circumstance for ranking.
It doesn't matter how special you or your mother think that you are - if Google search is important, you make changes that Google has made important.
Usability is key. One thing I noticed - the ads really don't seem clearly marked. They are marked, but not in a way that they are immediately apparent as ads. That's a huge no-no to me.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/5/13 1:38 PM|
I have nothing to say since
StackOverflow > Daniweb
simply because of no popup
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/5/13 1:41 PM|
> You'lls also found countless people that have offered you tips - most of which you disregarded. Do whatever you want Dani - but you can't claim that you haven't had loads of free help here.
I'm not claiming that I haven't had lots of help here. And I *didn't* disregard most of it. Most was implemented, actually. In fact, I even said earlier in this thread that I got a bucket load of help here, I implemented everything I could find to implement, and now I feel there's nothing more to ask from this community. Did you read what I posted?
And that spam thread about balloons has since been deleted.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/5/13 1:45 PM|
Lysis, we didn't have the popup for 8 months before we were hit. We put it back out of necessity after being hit because it increases our conversions by more than tenfold, and with ten times less traffic, we need to do *something* to compensate!
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/5/13 1:49 PM|
Nobody is asking for special circumstances Ashley. There are numerous UGC websites that were affected around the date she mentioned, which is being coined as the "Ghost Update", most a few days earlier that don't fit the norm for the types of sites that are being affected by Panda or Penguin. A Panda update officially happened on November 21, yet many UGC websites took about a 30% traffic hit a few days earlier with no official update from Google. I am not saying that UGC websites should not necessarily have ranking factors against it, but for such a large number of UGC websites to take a hit it would be nice to get an understanding from Google what happened on those dates since there was no official update that I know of. Was this an intentional update targeted at UGC websites? Or was this a side affect to one of their other tweaks that causes unintentional results? As a programmer I know how easy it is to introduce new bugs that cause adverse reactions when trying to implement features or fixing other bugs. So it would be nice to at least get clarification that they are aware what happened and that it was intentional, or they were unaware and are looking into it.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/5/13 1:52 PM|
> and ask for a special circumstance for ranking.
I'm not asking for anything. I'm not asking for special consideration, for help, or for suggestions. I thought I made that extremely clear in my previous posts in this thread. I even went so far as to previously state that I appreciate all of the previous help I received so much that I don't think there's anything additional to say.
I'm posting because I was encouraged to do so by some folks both on DaniWeb and on WebmasterWorld to shine some light onto the hit that quite a handful of UGC sites (myself included) seemingly took all at once back in November, that occurred just a few days before an announced Panda algorithm update.
Additionally, I'm opening up my entire Google Analytics (see attached Excel spreadsheet in first post of this thread) to anyone who was hit by either Panda or this UGC algo to give the opportunity for them to compare my data to theirs and work to figure out exactly what's been going on.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/5/13 2:07 PM|
Ghost update? The only thing more annoying than pop ups and tricky ads are people making stupid algorithm names.
Google has some large named algorithm updates - let us rely on Google to tell us what they are.
Otherwise, their algorithm updates HUNDREDS of times per year and low-quality UGC has never helped ANY user. There has to be enough good stuff if you're going to let your website fill with content that is potentially low quality.
Google never said anything about UGC being inherently bad - however, I would never expect them to rank a site well that has a bunch of bad UGC.
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/5/13 2:21 PM|
Many of the large named algorithm updates weren't coined by them, and instead regular folk which they adopted. It is just an easier way to identify what update we are referring to. Just like with software, many times a release has a certain name associated with it that is easier to remember, instead of having to remember 5.43.2112, etc. I am sure that is really more annoying than popups and tricky ads, can you exaggerate anymore Ashley?
Back on point, yes they make HUNDREDS of updates per year which is exactly why they may not be aware this problem exists with UGC websites if a bug was introduced. Or maybe they do, and it was intentional. Would be nice to know either way. They are not perfect in how they algorithms rank, they have acknowledged that. What leads to future changes in their rankings? Anything from internal reviews, human raters, or feedback from website owners. Maybe, just maybe it is possible that there is something to be fixed on their end? Or maybe not? Something happened in mid November and with Google trying to be more transparent with their information on Google and Panda updates, would be nice to know if there was some sort of other update that happened here, such as a UGC update, or whatever.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/5/13 3:02 PM|
Dani has had ranking problems for a long time - so no, I don't think Google accidentally caused her issues.
If you think there's something amiss, even without a manual penalty you could submit a reconsideration request.
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/5/13 3:15 PM|
And that is just your opinion. My opinion is that she was collateral damage because a good number of UGC sites were affected in mid November, not just her. Only Google could confirm or deny if there was some sort of "Ghost Update".
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/5/13 3:39 PM|
Ghost update ... :D
UGC has been a buzz word for some time, fraught with high expectations of a whole new "Web 2.0" - these times are long gone. UGC has proved to be notoriously low quality content, often erratic snippets of unreliable opinions and emotions instead of useful information and facts. Search engines necessarily have to keep their results clean from too much of it, whether they do this in "ghost" or "animal" updates is not important.
(I can't say too much about daniweb itself, because I'm already told by the initial question not to treat this or that or another topic because it would be considered "hating" the site. The whole thread is obviously again meant to blame a "Google bug" for all evils the site has been subjected to. Not thin content. user generated spam, excessive page numbers, deceptive ads, nasty popups, redundant content - nothing of that, no, it has to be a "Google bug". So be it.)
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/5/13 4:32 PM|
UGC has proved to be notoriously low quality content
You do realise the irony in that statement, right?
We are meant to be engineers, programmers, internet experts - this is a ridiculous directional turn. If you aren't interested just IGNORE US. It does more damage to any discussion then saying things like:
Where did anybody discussing the 'update between 16/17/18th of november 2011, that came between panda 21, and panda 22' mention anything other than support for Google in general?
Coincidentally I'm currently writing new filters that would love your comments.
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/5/13 4:47 PM|
>>> You do realise the irony in that statement, right?
I do *g*
Why? It explains why UGC isn't performing in the search engines that well anymore. There's no need to to look for bugs, special updates and what not in order to see where the general decline of UGC in search is coming from.
|hitchhiker||6/5/13 4:49 PM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/5/13 5:12 PM|
And look at your other post:
>>> We're really just asking for some simple
>>> intel/guidance that is specific to our situation
What I said about UGC is exactly that, a simple and straightforward explanation. I understand you're surprised by this "ridiculous directional turn", but you're just looking in the wrong direction, all the engineers, programmers, internet experts you mention must somehow be lacking the simplest SEO tool of all if they need bugs and updates and other fancy technical stuff in order to understand what's going on: common sense. Look at daniweb ... I wouldn't want to have too much of this stuff in my search results.
And if I was Dani, I'd also look at some metrics related to user satisfaction, like: bounce rate (Alexa has it at 80%), time on site (Alexa says it's as low as a single minute), pageviews per visit (Alexa has less than 2) and percentage of returning visitors (most likely not too many) ... she may be in for yet another surprise of the kind you experienced with my statement ...
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/5/13 5:22 PM|
Here's yet another thing I'd be quite uneasy with:
Daniweb.com’s Regional Traffic Ranks
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/5/13 5:44 PM|
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/6/13 1:13 AM|
Wikipedia is a UGC site, but the quality is extremely high. It's rankings are still extremely high. See the connection?
So UGC sites haven't been "picked on", low quality sites have been dropped lower in the rankings.
Dani, I know you think you have done everything, but I go back to my original points:
1) Linking to thin content from the homepage - the first link I saw on your homepage was to this page: http://www.daniweb.com/software-development/vbnet/threads/456006/excel-reader - which feels very thin / low quality. Can you link to high quality content from your homepage?
2) At the bottom of that page there is a tag section, there are about 70 tags (way too many) and most seem point less, e.g. if, how, to, etc, etc.
Matt cutts has said he doesnt like tags, its creates a lot of duplicate pages, so I would delete them all, it feels radical but to break out of panda you need to do radical stuff.
I know Dani said "Our tags are our primary means of navigation on DaniWeb." but how can this tags be of any use?
it is totally useless spam. Amazon seems to rather well without tags so if tags are your primary navigation there is something wrong with your navigation, search should be the primary navigation for a site with 422,000 pages in google. Also if you must have tags there need a massive cull in the example I gave you have 70+ tags, now look how stack overflow do tags - http://stackoverflow.com/questions/16956536/grid-showing-empty-rows-extjs - only had 1 tag
3) I hear that you noindexing thin threads. I would go a step further and 404 thin pages. Maybe do a review and if 99% of threads don't get a reply in the first 30 days, simply delete them automatically after 30 days. Or something like that. Matt has said that you should 404 low quality page, noindex is not as good
4) You got 422,000 pages indexed in google, are they all high quality? If not can you delete the low quality ones?
I hope that helps.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/6/13 2:43 AM|
Maybe do a review and if 99% of threads don't get a reply in the first 30 days,
That's a good suggestion, though often somebody comes in via Google and answers it. It's a tough call, but I suppose (considering) that's probably the way to go now. I'm off to implement this.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/6/13 2:51 AM|
Implemented as: If a thread is over 2 days old, and nobody has replied - it's 'no-index'. Thanks.
|Re: DaniWeb hit in November '12 and haven't recovered since||Marcus S||6/6/13 3:28 AM|
better to use 'if there is a response, index' than 'no response in 2 days =noindex' or you will get all the pages indexed even when poor quality and have loads of 404s for years to come - once google have found a page it takes them a long time to forget.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/6/13 5:20 AM|
99.9% of our threads get answered within a few minutes, the rest get deleted normally (if they're just nonsense), this won't be a problem - it's mainly for the few that slipped through over the years.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/7/13 3:59 PM|
My apologies for coming off rather harsh earlier in this thread. I've had so many discussions upon discussions about everything I've done and changed and why and how and whatnot, that I really just didn't want to rehash everything all over again and re-explain things that I already went into great detail on in other threads in this very forum.
However, I have a bit of time right now on this cold, rainy Friday to go into greater detail, so I'm going to attempt to do a better (and more thorough) job at replying to everyone.
Before I get started, however, I want to mention that I went into GREAT DETAIL exactly how and what was done in the WebProNews video I did at http://videos.webpronews.com/2011/05/daniweb-speaks-out-on-recovering-from-google-panda/, and everything that wasn't mentioned in the video was mentioned in my other threads.
Luke => even pages that have 100's of words can be low quality thin content - keep in mind part of the algorithm tweak was to nock out scrapped content , where pages of junk text were used to fake the ranking.
We noindex most member profiles (except for those by our top members, because they appreciate being able to have their member profiles found by people Googling their name, much the same way as when people google your name you might like for your Facebook, Twitter or LinkedIn profiles to appear in the results). We noindex most navigational pages (everything except our homepage and top level category homepages) because search engines want to send people to content pages, not navigation pages. We also noindex all of our tag pages and we disallow via robots.txt our actual tag clouds themselves. And then, of course, we noindex all forum threads that have 0 replies. As with all well-moderated forums, nonsense and spammy content is quickly deleted. We chose to use the 410 http status code instead of 404 to indicate that the page did exist at one time but since has permanently been deleted, as opposed to 404 which is simply a response from the web server that the file doesn't exist anywhere for unknown reasons.
terry => Hi Dani, congratulations on the 1 million Plus members (WOW). And sorry about the extra pounds :(
Thank you. :)
terry => The second issue (from a new user point of view) was "start a new community discussion" that only displayed the login drop down after I had edited the title and chose the forum and started to edit the editor.
This is by design. I didn't want to scare off someone with a huge form right away, but I also wanted to make it a one-step process as opposed to having to having to register on a separate page.
terry => The title "pop ups" on your navigation are useful, but don't you think you loosing out on valuable optimization. Text on the page carries more weight than in a link title (my opinion).
It looks too cluttered if the descriptions of all the threads always appeared on the page IMHO. I like the way they appear as tooltips instead. I think it makes it easier to navigate. Thank you for the feedback though. I think it's just a matter of taste, and I'd rather it be more usable and be a cleaner interface for the end-users than try to optimize for google.
julianhh => Matt cutts has said he doesnt like tags, its creates a lot of duplicate pages, so I would delete them all, it feels radical but to break out of panda you need to do radical stuff.
Ashley => You'lls also found countless people that have offered you tips - most of which you disregarded. Do whatever you want Dani - but you can't claim that you haven't had loads of free help here.
Where on earth did you get the impression that I disregarded everyone's suggestions? Can you show me where I claimed that I haven't received any help??
Ashley => Sure - spam/UGC is one thing. But Dani has lots of opportunities to improve her site and has had countless hours/help given to her. If she chooses to implement, fine. But you can't go through and said you did 1 thing - ignored 20 - and ask for a special circumstance for ranking.
Where did you get the impression that I only changed one thing and am now asking for special circumstances??
Lysis => I have nothing to say since StackOverflow > Daniweb simply because of no popup
In the thread http://productforums.google.com/forum/#!searchin/webmasters/daniweb/webmasters/MRgEeIe0Nyc/QMXCms5VJ5QJ you made the following comments:
While I understand that you REALLY hate our welcome modal, it only appears ONCE FOREVER until/unless you clear your cookies. Certainly the merit of a website as a whole can't be *entirely* in the balance of whether or not it has a single modal ad that only appears once. I can understand if it keeps appearing. I can even understand if it is frequency capped to appear once per day or once per week. I can even understand if it's a pet peeve of yours. But making it the end all be all and thinking that it makes the site go from an A to an F just because unregistered users get a welcome modal once forever that tells them what the site is about and encourages them to join I think is a bit much.
> Ashley => Dani has had ranking problems for a long time - so no, I don't think Google accidentally caused her issues.
We were hit by Panda in February 2011 and then recovered in July 2011. We were then hit by "something" in November 2011 and we haven't recovered since. Why do I think that it was "accidental" on Google's part that we were hit again? Partly because Matt Cutts said so :) About two months after our July recovery, Matt Cutts did a live Q&A session in which he said: "We’ve made many, many changes over the last few months, even within Panda, trying to iterate, find new signals. You see a site like DaniWeb complain and then we find signals and say, okay, here’s a way we can differentiate between this site and the sites that might be a little bit lower quality." So basically I took that to mean that Matt and the Google webspam team realized we were penalized unintentionally and deliberately worked towards making Panda not hurt us. (http://www.youtube.com/watch?feature=player_detailpage&v=R7Yv6DzHBvE#t=1113s)
Ashley => If you think there's something amiss, even without a manual penalty you could submit a reconsideration request.
It was my understanding that reconsideration requests are intended only for manual penalties. If it is, indeed, an algorithmic bug, the support folks answering the reconsideration requests wouldn't be the right people to reach out to. I'm hoping that the right people are reading this thread :) *wink wink*
Bigwebmaster => My opinion is that she was collateral damage because a good number of UGC sites were affected in mid November, not just her.
What's very interesting is that, from what I am aware, all of the UGC sites that were affected in November are either on custom platforms or have very out-of-the-box skins that would make it not clear to a bot, especially at first, that we are forums. For example, we don't look like the majority of vBulletin's, and InvisionBoard's, and phpBB's out there. Now this is just an [educated] guess, but I'm guessing that it's built into Google's algorithm to give forums a bit more leeway when it comes to judging the quality of content than editorial sites, at least on the grounds of it being understood that forums are less likely to be grammatically correct and that type of thing. (I'm not intending to get into a debate as to whether Google thinks forums are as useful or can have content as good as editorial sites, because that's not something I can say I know the answer to.) But when a forum isn't recognized as a forum by googlebot, perhaps we were the unfortunate bunch that got classified wrong and are having our content being held to editorial standards instead of UGC standards.
This theory is actually supported by what I wrote in my original post in this thread:
This supports the theory that Googlebot was not sure what it was looking at when crawling our site, and potentially misclassified us.
luzie => I can't say too much about daniweb itself, because I'm already told by the initial question not to treat this or that or another topic because it would be considered "hating" the site.
I didn't intend to come off as elitist or not open to constructive criticism. I just don't want to rehash the exact same suggestions and discussions that I've had to exhaustion already. If you have new suggestions or constructive criticism for me, then I welcome those.
luzie => The whole thread is obviously again meant to blame a "Google bug" for all evils the site has been subjected to. Not thin content. user generated spam, excessive page numbers, deceptive ads, nasty popups, redundant content - nothing of that, no, it has to be a "Google bug". So be it.
I already have 3 different threads in this forum where I basically opened myself up and asked "What am I doing wrong?!" and I had hundred-page long discussions about thin content, and spam, and ads, and popups, and all those things. IMHO, I fixed all of the problems that *I* felt were wrong with the site, I explained in great detail what I changed, and I changed A LOT.
Now I'm trying to open up an honest discussion saying, "What if it's a google bug?!" I'm presenting my opinions for thinking so, and was asking the community if anyone else thinks my theory has the potential to be true. I'm not bitching and complaining that Google is fucking me over. I think that I'm presenting a rational potential possibility, opening up my entire analytics, and saying "Have a stab at what I'm seeing and see if it makes sense?" IMHO I'm allowed to open up a discussion in this forum without being labeled a complainer asking how to get more traffic.
luzie => percentage of returning visitors (most likely not too many)
More than a third of our visitors are recurring according to Google Analytics. I am unaware what the industry average should be, but I was told by an extremely well known SEO who I let log into my analytics that my numbers were not in line with what he was seeing from any other clients of his who were hit by Panda. Once again, something to make me ponder that perhaps this isn't Panda after all.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/7/13 4:05 PM|
I just want to commend that this was a typo. We were hit in November 2012, not 2011. We lasted well over a year since being hit by the original Panda, and were hit about six months ago in which we haven't recovered.
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||6/7/13 10:38 PM|
Hi Dani I would like to comment on one of your posts.
I total disagree with you noindexing navigation pages. The spider needs to crawl navigation pages You should Noindex, follow those pages so that the spider is not "interrupted", and can crawl all navigation links on your site.
That could also explain the message in webmaster tools about your forum. I think the spiders battling to crawl your site.
Problem solved, do I get the best answer?
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/8/13 3:10 AM|
Maybe forums are viewed in a different light , but maybe they arent.
Here is the cache text view of your site.
You have a really thin front page when its just text.
Also the description is too long and the title doesn't really get honoured with content.
As thin content is always going to be an issue you would probably want to have a very strong front page.
This is the essence of the problem.
Looking at one page thats indexed
It has "String" in the title , but not in the content. And the meta description is the intro from the post , which doesn't honour the title or the important parts of the post.
So then to an algorithm this looks like its a published article by someone who has no idea about what a search engine is looking for. And thus gives it a low quality ranking.
And this is the content your site is fuelled by.
Site development is supposed to be just about making a great site for your users , but in essence the forum is a site made for you by users ?? In content terms of course.
Maybe high ranking forums need to have all content manually allowed to be indexed? to maintain the title of an editorially controlled post
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/9/13 6:42 AM|
> I total disagree with you noindexing navigation pages. The spider needs to crawl navigation pages You should Noindex, follow those pages so that the spider is not "interrupted", and can crawl all navigation links on your site.
Terry, I was under the impression that when you add the noindex meta tag to an individual page, the 'follow' part is implied. It does crawl the page but simply doesn't index the individual page into Google. Am I incorrect?????
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/9/13 6:54 AM|
Luke, that's my point. All of the qualities you brought up are in line with that of a forum:
Our homepage is a listing of latest articles.
Main article listing on Xenforo => http://webcache.googleusercontent.com/search?q=cache:xenforo.com/community/forums/suggestions-bigger-features.9/&rlz=1C1CHFX_enUS528US528&strip=1
Main article listing on vBulletin => http://webcache.googleusercontent.com/search?q=cache:www.vbulletin.com/forum/forum/vbulletin-sales-and-feedback/vbulletin-pre-sales-questions&rlz=1C1CHFX_enUS528US528&strip=1
Neither seems to be any less thin to me than we are.
But that's the nature of forums. People make up their own titles and they make up their own descriptions. I believe there was another thread somewhere on these forums about it being a very fine line between editing post titles to make them more descriptive and editing post bodies to fix their spelling/grammar, and giving googlebot what it wants, without coming off like thought police or having it be detrimental to the community that we have too tough a moderation team.
That's the nature of UGC forums though: All published articles *are* written by someone who has no idea about what a search engine is looking for, and they are not intended to.
|DaniWeb hit in November '12 and haven't recovered since||terry noname||6/9/13 9:29 AM|
I wouldn't bank on W3c's statements "subsidiary links to be explored".
Its a known fact search engines interpret noindex meta tags differently, some even disregarding them.
Take note what Google say.
"When we see the noindex meta tag on
a page, Google will completely drop the
page from our search results, even if
other pages link to it."
That's what they mean by crawling the page (to come across the meta) The page is then gone.
But with "noindex, follow" you don't block the crawlers completely. This way would allow link equity to flow through to the pages.
And here they don't mention following links.
"To prevent all robots from indexing a
page on your site, place the following
meta tag into the <head> section of
<meta name="robots" content="noindex">"
Youtube video - When would someone use "noindex, follow" in a robots meta tag.
Matt Cutts blogs on noindex meta tags (a poll)
In his blog he does backup crawling links on pages that are "noindex, follow", and there is no mention of them following links with "noindex" meta.
Here I go again (the tag didn’t show in my previous post)
Tell us about a middle ground. In the case:
Google to not show that page at all, but it still following the links on that page.
meta name="robots" content="noindex, follow"
Harith, I just got to your follow-up comment. In that case, currently Google wouldn’t show the page but would follow the outgoing links.
I haven't come across a direct post by Google that state they follow links when the page has "noindex" in it.
If you have a link from a Google representative stating they crawl links with "noindex", please share it with me.
I really don't see the point in them coming back to crawl the links if you've told them to drop the page from search.
Because for all you know, you could be blocking out important links that would help the spider in crawling your site (navigation pages, which in my eyes are important for the spider)
Just my way of thinking (unless I'm missing something)
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/9/13 1:37 PM|
Site still isn't great quality. Winking doesn't increase your charm.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/9/13 4:18 PM|
Luke, that's my point. All of the qualities you brought up are in line with that of a forum:
I've used Dani's site over the years because the forums contain a tremendous amount of valuable Q and A information that actually solve answers to questions. Might as well say that Sitepoint isn't great quality either, nor any other site that helped most of us learn how to program and get answers to our questions. Unfortunately one of the least useful forums I've found is this one, right here. The truth is that it's mostly speculation, not that I can blame anyone specifically, because Google only provides the broadest descriptions about what it's really doing behind the scenes. There is nobody to officially explain how some very odd results might occur so that webmasters can properly and cost-effectively deal with cause and effect.
Dani brings up a troubling issue that many SEO's here aren't sensitive to as their focus is to deconstruct a site and may not have been involved in building it for many years. Google's algorithms are thought of as progress but are also having the effect of reducing a decade's worth of work to rubble not because it sucks or it "isn't great quality." If you've spent 10 years of your life trying to bring programming answers to people and Google suddenly changes the rules of what they think they want, should you have to hire an SEO for $10,000 to guess what the problem might be (which is always speculation, some more likely than other guesses.) Then there are the developers necessary to deal with all the new rules to reprogram the site for what Google now wants today for what it didn't want 5 years ago. The net effect is that if this is the case, many great UGC and venerable sites will continue to go out of business, not that this is a business concern to Google.
FYI - I've never met Dani personally. Like Dani, I run a large site that was hit by Panda. We didn't see any recovery effect of what SEOs identified as "sure fire signs of Panda that require correction." Only one thing made a significant difference (I am assuming) but I won't state that here as it will set off a firestorm of discussion. With the latest algorithm change our 12+ year forum and site went to zero PR. Scrapers have a higher PR than we do for our own content. No, we don't sell links. The site isn't perfect but there is absolutely nothing anyone could identify that would knock down the buiilding.
The reason I went to the public with my site was because before I'd consider spending money on additional SEO help, I assumed that most SEOs would be able to agree upon an area of need they could see. Then I'd be able to know how and where to spend the money on the help. Unfortunately it has been mostly speculation. Quite frankly I don't know if we'll be able to get an answer unless it's from someone at Google. I'm hoping that we'll finally get to that point before it's not too late.
I'm sympathetic to the SEOs here too understanding that 90% of the time there are very inexperienced people here posting about how unfair it is that what (usually spam) ranked yesterday no longer ranks today. Frustrating to hit "paste" all the time and make it through the cries about how unfair a high ranking affiliate casino site was now finally caught. But people like Dani are educated, professional, hard working people who are easily on par with most here in terms of knowledge. She has legitimate content, whether or not Google likes that flavor of the month today. It certainly has value. And the problem most of us are facing is that now 90% of our efforts are on trying to figure out what Google doesn't like and 10% on building content.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/9/13 4:30 PM|
Her site was good until she got greedy. Now she is outranked by people with nicer sites with better quality answers (stack overflow). That's just life.
I always bounced from her site with the popup and I know logs of people who did.
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/9/13 4:37 PM|
Well I thinks it's a case of "you can't have your cake and eat it as well"
I see it as , you as the webmaster are responsible for what posts are indexed - if they are thin it's probably best to not submit them for indexing.
If you want to be ranked for terms your users type as questions it may only be beneficial to the search results if they are well put together posts.
I recovered my site from Panda , and my rule that I made for my self was to remove or condense any pages bellow 600 words - then those 30 pages I printed out and reviewed them with a red pen like a college assignment .
If I ran a forum and I wanted to be ranked I would personally switch the posts to index - and maybe even jump in on some threads to clarify or fill in gaps
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/9/13 4:38 PM|
Stack overflow has 12 million dollars in VC to recoup. [http://on.mash.to/h29Yvm]
Corporate route, so best of luck to them, I also like their site.
Dani's independant forum contains information I'd like to get at, but I can't now. Despite all this back and forth, you've not shown why her site should have been thrown away.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/9/13 5:09 PM|
Dani had her chance. She went greedy and didn't listen to her users.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/9/13 5:18 PM|
Wow, the vitriol with which you seem to approach her concerns is staggering..
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/9/13 5:24 PM|
Hello friend! U r very observant. Thank you for your feedback.
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/9/13 5:38 PM|
Passion could easily be taken as vitriol :)
We are all nerds after all
Moving on - search seems to be heading to branding , thin front pages miss out on branding.
With the notion that everything you have done up until now has gotten you to this point. And it's not where you want to be.
Maybe looking at it from a point of a 'relaunch' - take a giant step back and see how you can reinvent your approach .
How would you view the situation if you were launching the site today for the first time - with all the content you have - if you can see it from that point if view it will be with new eyes
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/9/13 5:47 PM|
Luke, respectfully, I don't understand you're trying to suggest. What is "branding" that you're referring to and how does or should that make any difference with regard to search engine results? What exactly is there to reinvent? If I was going to relaunch her site, I might change the design a little but it is pretty easy to navigate to find what you want, no?
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/9/13 6:18 PM|
A Brand is an Authority - its an entity that is greater than the sum of its parts.
The rich sites that fall from the ranks - would instantly get to work of brand recognition.
Relaunch - if you were to relaunch a site that you are unhappy with in the same way, you would still be unhappy with it :(
Trying to reinvent the site to a way that allows it to rank - thats what we are talking about right? If something doesn't work as you wish a new approach is required. What I was leading to is ...
Go with me on this (imagination required) - Its relaunch day , You have the site setup , you have 100,000 pages , a good portion of them are not suitable for a search engine. A good portion are going to let you down if you let them get indexed. Do you go through what you have got , and select the top 1000 to index , thus making the indexable content all good stuff.
Or do you keep doing what your doing and put the whole lot in the index , thus weakening your brand and rank
The hardest part I see with people who once ranked and got heaps of cash from ads, is that when they fall, they have all the baggage of how it used to work - thats the biggest sticking point I see across all platforms on this forum.
Each time Google reinvents their search algorithm and it kicks you out - you need to reinvent your self
If you can truly see your self as doing a relaunch , and get in the midset of starting again you can start afresh on a sold platform.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/9/13 7:08 PM|
Luke - thanks for the extended explanation in trying to convey what you mean. I understand what branding and relaunching mean but what is the course of action you're suggesting? In short, it's the same thing I was alluding to -- blow it all up and see what sticks. As a webmaster, we've invested too many years and absolutely need and should have a set of rules we can follow so that this isn't a hugely expensive game of pin the tail on the donkey.
Branding - It should not have any effect on an algorithm. A brand effects how people feel about a product - people might buy something with a hot brand whether or not the product is actually crap. :) Google's algorithm is supposed to sort the wheat from the chaff and show no bias in its rating what it sees solely on a brutally rigid set of rules in binary.
Removing 99% of a site's content and indexing 1% - Scares me to hear this. Not only doesn't Google work this way but it's impossible to do what you describe. How does a site owner qualify posts Google is supposed to be looking for? People love my site. It has lots of legal answers that people take time to craft for each person's situation. How can I even choose 1,000 of the "best" Q&A out of my site on a myriad of topics, among hundreds of thousands of posts, many of which would apply to a specific situation? In many respects there is no "best" answer, only a best answer that fits that situation. With Dani's site you're talking about a similar issue, e.g. code examples that people will post. How can you choose your "best" jquery issues to feature? You can't. It's a repository of information that Google is supposed to know how to index and navigate better than anyone else and provide good results and we do what we can using best practices.
Cash and Deindexing Results: Many of us didn't rake in "heaps of cash" from ads. And to deal directly with your point about deindexing content, I didn't realize I had an attorney directory with over 200,000 "thin" pages that were in a subdirectory. Panda came and whacked us to a noticeable degree. Can you guess what happened when that entire directory was removed from Google's index outright? Your argument, logic and many others suggested that some noticeable return was imminent for sure. My advice is that unless you are darn sure that you should deindex content, you're getting yourself possibly into bigger trouble. As I said, only one thing worked that I don't want to discuss here. Not a full recovery but was the only thing that began to improve results.
This is a nice euphemism you provide to someone who is down in the dumps and whose career has passed them by. It doesn't provide any direction though. So what am I or Dani supposed to do now? Spend $25K on stuff just to blow it all up and see if Google likes it? The worst part is that NONE of that money will be spent on the good content that Google is supposedly craving. So now content is not king - content is worthless unless it's presented exactly the way Google wants.
I can understand that some of you aren't fans of the way Dani presented her content and had issues with her pop-ups. Fair enough - I can't speak to that matter or others which Lysis might take issue with.
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/9/13 7:38 PM|
What if reinvention is all that you have left?
It doesn't need to be scary or uncertain - it can be exciting , adventurous and empowering , the choice is yours.
Thomas Edison took 206 revisions of the light bulb before it stayed lit.. the whole time his wife and associates were nagging him , putting him down ... And we all know how that turned out.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||6/9/13 7:52 PM|
Chalk me up as someone who has moved on to stack-overflow and led-bulbs. Simply better with less annoying side effects. More light and less heat.
Google has taken the place of our stars.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/9/13 8:00 PM|
If reinvention is all you have left then it's time to get a new business! From what I recall, Edison was able to afford tinkering around with the light bulb after he struck it big...
Regarding stack overflow, it looks and acts exactly the same as a forum to me. It's just content presented in a different way that, personally, I frequently find much more confusing and very messy at times. And that's unfortunately the point. The content isn't different or any better, it's just the container. And unless you've got a team of developers, it looks like the small to mid sized business will never be able to afford what it takes to keep up with making whatever it is makes Google happy this week. I certainly hope it's not the case.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||6/9/13 8:47 PM|
No. It's what makes me happy.
And the LED light-bulb is just a filament in a plastic package.
"All our final decisions are made in a state of mind that is not going to last." -- Marcel Proust
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/9/13 8:58 PM|
LOL. A good examples of bad results is this discussion. As per another I have, someone keeps ranking my posts as negative. Doesn't matter what it says. Obnoxious posts by some "high levels" (not talking about you guys) somehow get up rankings. Now might we be seeing some type of collaborative effort between posters? Hmmm... I wonder. But Google really doesn't care to administrate. The machine will roll on until some competition or visible issue causes someone to take heed of an issue..
|Re: DaniWeb hit in November '12 and haven't recovered since||Richard Hearne||6/9/13 11:01 PM|
>> Search engines necessarily have to keep their results clean from too much of it
Nearly every technical query I enter into Google returns UGC sites in the top positions...
Quality is a subjective measure, but one thing I do know is that Google still favours many UGC sites.
|Re: DaniWeb hit in November '12 and haven't recovered since||Richard Hearne||6/9/13 11:05 PM|
>> Terry, I was under the impression that when you add the noindex meta tag to an individual page, the 'follow' part is implied. It does crawl the page but simply doesn't index the individual page into Google. Am I incorrect?????
Yes. Terry's comment above is nonsensical and should be disregarded. Unless you set a page to nofollow or block access via robots.txt all links are followed by default.
|Re: DaniWeb hit in November '12 and haven't recovered since||Richard Hearne||6/9/13 11:07 PM|
>> Just my way of thinking (unless I'm missing something)
You're missing something Terry. Deafult is follow unless otherwise directed. Noindex does not change this.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/9/13 11:23 PM|
Each time Google reinvents their search algorithm and it kicks you out - you need to reinvent yourself
I don't think that's what Google intended, they'd been destroying any chance of a sustainable business on the net. If 'product' sites were treated like that, you'd see the end of e-commerce. Google, themselves, would not survive that kind of business environment.
I think you're all trying your best to explain away her problem in terms that you can understand, and have access to. You are not willing to accept that Google corrects itself, and this may be partially something needing correction.
As for dani vs Stack overflow -> Yes S.O is a great site, with corporate backing they don't need to display ads, they have an interesting format for tech questions.
That shouldn't be the only choice in tech q&a; Sometimes we like to see 'vertical flow' discussion too.
|Re: DaniWeb hit in November '12 and haven't recovered since||alistair.lattimore||6/10/13 6:22 AM|
You're correct, pages are follow by default - else Google wouldn't be able to crawl the majority of the internet.
As such applying "meta noindex" (without explicitly adding follow) is actually "meta noindex, follow" by default.
Assuming a given URL isn't blocked by robots.txt, Google's behaviour when crawling "meta noindex" is to deindex the specific page, flow PageRank through all links except those with a rel="nofollow" attribute and follow the links within the page for discovery of other pages.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/10/13 6:47 AM|
I'm going to leave my noindex meta tags as is without explicitly adding the follow part too for now. I think Terry got confused by the difference between spidering, crawling, indexing and caching. On my thin pages, indexing and caching are bad, but spidering and crawling are good.
Regarding reinventing ourselves ... well, we did. We were originally hit by Panda and made a billion and one changes and then we recovered a few months later. However, in my eyes, they were all SEOish changes that were all just band-aids to cure our google woes. That's why, soon after, I began rewriting the entire platform that powers DaniWeb from the ground up. While the branding, color scheme and overall aesthetic remained them (for branding reasons), the entire presentation of information was completely rewritten, the site became LEAPS and BOUNDS faster, became much more AJAX based, and *really* made a huge impact to the quality of the content being posted. One of the biggest things we did was switch from BBCode to Markdown, and put a lot of safeties in place to ensure that everyone formatted their code correctly. Previously, it was a huge drain on our moderation team to manually wrap posts containing code in code tags, posts containing quotes in quote tags, etc. because they were being too lazy or didn't know how to do it themselves. Now, our system makes it a requirement and formatting in general is much easier. In fact, there was such a night and day difference in the overall presentation of UGC content that our moderation team went back and manually fixed 80,000+ posts over the course of a few months that were incorrectly formatted by their posters from the old system. They were never modified back in the day because there were just so many of them that our moderation team was overwhelmed. Now, aside from Stack Overflow which also uses Markdown and does something similar, we no longer suffer the issues of every other programming forum out there that still uses BBCode (a lot of my friends also have large programming forums and they all agree that incorrect use of bbcode is their #1 pet peeve). Not us anymore!! Now both humans and googlebot alike will be able to distinguish where the code blocks start and end, where the quotes start and end, etc.
The entire reinvention project took about 5 months and was done at a time long after we had recovered from Panda and survived a good handful of Panda data refreshes and updates, when we had the highest traffic we've ever had up until that point. I felt pretty confident in my Panda success at that point. It was done because you don't have to wait for things to be bad before you should think about improving them. Even though we had recovered from Panda and were doing great, I personally saw room for improvement and so I jumped on it.
I carried over all of the SEO improvements I made during my bout with Panda into the new system.
Fast forward to today. There are two things that I think could be killing us right now.
#1) We're now on a proprietary, unique platform now, and perhaps googlebot just doesn't understand / recognize it as a UGC forum because it's not like any other forum system they explicitly know about such as vBulletin, etc. Our content has only improved since our recovery, and yes, Panda is constantly adding in new signals and such, but I think it's a bit coincidental that we were hit, we recovered, Matt Cutts all but said that we were *not* a low quality site in Google's eyes in his youtube video, we make it through a whole bunch of panda updates and data refreshes, then we switch to a new platform, are suddenly hit again, and then have Google webmaster tools telling us we're still using vBulletin when we're not (proof that googlebot doesn't understand what it's looking at).
#2) While I did carry over all of the SEO improvements I made on the vBulletin platform during my bout with Panda into our new system, there is one change that I made. I changed the URL structure of our threads to include a slug of the thread title in the new system. Everything is 301 redirected, of course, but we're essentially 301 redirecting 10 years worth of accumulated backlinks, and they say that you do lose a tiny bit of link juice with every redirect you do. We were clearly hit on a specific day though, so it's definitely an algorithmic penalty that got us, as opposed to just a rolling change in our backlink profile.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/10/13 8:01 AM|
Dani, you might find this interesting: http://www.webmasterworld.com/google/4582387.htm
Your story is almost identical to ours and many others. We either get a 'fix' or it looks like we'll slowly fade away. It doesn't seem to have anything to do with 'keeping up with the times', quality, or user interaction levels - those, as we know, are always on going. *it seems so far* this has randomly affected large sites, with over 8 years on the books.
|Re: DaniWeb hit in November '12 and haven't recovered since||SGT Oddball||6/10/13 8:30 AM|
The 8 years thing is wrong, I lost 160,000 visitors per day from the 17th November ghost update and my site it only 7 years old, I did compare it to daniweb an alexa a few days ago (I know it's not the best metric) and the drops are absolutely identical since the November update so I'm pretty convinced whatever Dani is being penalized for I'm getting the same.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/10/13 8:38 AM|
@John, it's best to speak over at webmasterworld - this forum isn't really getting us anywhere. Could you post there, and we'll collect info incase we get a chance to directly speak to the optimisation team?
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/10/13 9:07 AM|
For us it was June and August 2011. After that period we ended up spending ridiculous amounts of time trying to make changes to no avail. As I said, only one thing worked and showed noticeable growth. I will be re-implementing this "fix" once again. If I see more newfound growth, I will have something substantial to say.
When I talk about PR of our site, I guess I better clarify and say that the guesswork has corresponded with where our site has traditionally appeared in SERPs overall. For example, back in June 2011 one of our top 10 keywords dropped 90%. That's crazy. And as I've mentioned, I'll show you horrendous sites that continue to gain. I don't know that it is UGC sites themselves that are being penalized exclusively but there are some who I've noticed gaming Google in other ways and capturing visibility.
For example - one of the top 5 sites in my niche started (and still does) publishing RSS feeds as their own articles on their site, many of which have absolutely nothing to do with the niche. My search on hockey led to articles on this site, which was a nice looking feed for USA Today or something of that nature. Google was indexing their content and sometimes showed multiple results. The result came up whether or not I was logged into Google. This was the first trend I noticed which many were doing.
When I did a search on items related to the NFL draft, I noticed that some sites like Bleacher Report had at least a half dozen results in the first two pages. I'm not saying they gamed the system but Google was indexing the same articles in several ways, e.g. tags, rss feeds, etc. As a result, I was guessing that my pages weren't being seen because many of the positions were being dominated by just a few sites. Searches on common terms that I ranked for led to 2-6 entries for individual sites on the first five SERPs.
As a result of vBulletin putting an end to their best forum software product, we'll be moving to another platform as well. I'm glad that I didn't pull the trigger yet because I've noticed that stack exchange type sites seem to be performing with greater superiority than others. Dani's statement about her experience seems to echo my experience and concerns. This leads us back to one very huge issue as a content site that has been around since the 90s. The game is now much about form over substance. There is no such thing as "reinventing yourself" any more than you can tell a building owner to knock it all down and rebuild every few years. You can take up yoga or zumba if you want reinvention (and losing 5 pounds). But nobody can afford to keep recreating their CMS every few years along with all the work that goes into redirecting all the links.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/10/13 10:36 AM|
John Fairbrother, I posted my exact Google Analytics stats in the first post of this thread, so no need to compare using Alexa.
That being said, I'm becoming more and more convinced that there was some penalty that went out to forums and UGC sites specifically.
Check out this post on DaniWeb:
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/10/13 12:07 PM|
So what would you do if a Googler suddenly chimed in and said "Yes, I confirm we've rolled out the Ghost Update in November 2011" - would that help you in any way? And no, everything that is coming up here doesn't even faintly look like "a penalty", all I can see is perhaps a little algorithmic tweak of ranking logics regarding low quality UGC - nothing the users seem to have the slightest problem with (sockpuppet and buddy opinions excluded). I've seen several comments now on stackoverflow being way better than daniweb, so all this seems to be nothing but a little UDC - a user driven change.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/10/13 12:46 PM|
On Monday, June 10, 2013 3:07:49 PM UTC-4, luzie wrote:So what would you do if a Googler suddenly chimed in and said "Yes, I confirm we've rolled out the Ghost Update in November 2011" - would that help you in any way? And no, everything that is coming up here doesn't even faintly look like "a penalty", all I can see is perhaps a little algorithmic tweak of ranking logics regarding low quality UGC - nothing the users seem to have the slightest problem with (sockpuppet and buddy opinions excluded). I've seen several comments now on stackoverflow being way better than daniweb, so all this seems to be nothing but a little UDC - a user driven change.
Luzie - I completely appreciate your point. Many of us come here upset of huge swings that seem unexplainable, we vent, etc. A site owner must accept that perhaps we aren't as good as we'd like to think we are. Point made. The problem is that if traffic is a zero sum game -- and we at least qualify to be somewhere in the game -- we shouldn't be watching significant amounts of our traffic go the way of those who should never be qualifying. I'm not sure what makes stackoverflow "way better" than daniweb to the point of saying that, e.g. 30% of traffic should suddenly be diverted away so I'll drop trying to qualify either site's content. But do we know that this traffic went to stackoverflow and "superior" sites? I remember Dani's first drop and mine in June 2011 was ridiculous -- so I watched who gained. Dani can speak to her niche and I can only speak about my own for an example in which the right results didn't happen all too often.
For approximately a year (per quantcast, compete, etc.) let me show you just one site whose traffic grew exponentially and it looks much better today than it did then -- http://oz.vc/9. They are still ranked in the top 5 on page 1 for "la w for um" and have a laundry list of major issues that would make any SEO's head spin (and heart race if they were willing to pay your fees! LOL ) No navigation across the site, blatant link exchange, dozens of keyword heavy links solely for search engines. You would have enjoyed the 3 large rectangle Adsense ads in the header they had for many months - a sign of quality therein. No matter what I did, I had to watch them feed RSS into the forum, have multiple listings on the first few SERPs for keywords I used to rank for and no matter what I did made no dent in their rise whatsoever to the point where their traffic appeared to best mine. Nobody confused our two sites.
My point is as follows - if we try our hardest to play by the rules and we reasonably qualify for an "acceptable" site, we shouldn't be beaten badly by sites that flout what are supposed to be the rules and the "no brainers" that will allow us to rank. Scrapers should never rank above our own sites (as happens to me frequently), especially when Google indexed us before and must know what a ton of duplicate content coming up on another domain obviously means! Google may honestly believe they are making good faith efforts improving the algorithm. But (to quote a friend) it's like a giant foot coming down but not bothering to wonder how the multitude of ants and other small but important beings are faring. If Matt Cutts' rules are only guidelines don't have a reasonably good reproducible effect, we're not going to be able to adapt our websites properly and do the things Google says they crave. As it stands right now, we have no way to be able to understand how that which should move the earth to some degree may make no impact whatsoever. The fact that some sites do see major changes means that there is some truth to the guidelines. As to why some issues occur (e.g. scrapers > content creators), we really need some meaningful dialogue with someone at Google to help make sense of the very large gray area... and quickly.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||6/10/13 1:23 PM|
Re: Replaced by things like 'stack exchange' -> I don't have them in my sphere (thank god) -
* I've not really lost much, I'm still at 120k visitors a day.
* A LOT of my competition have disappeared, they had as much 'quality content' as any of us who survived.
* Another site has been replaced by a domain with no content, no backlinks and a single google places reference.
* Things are broken, discussing anything here just feels like 'waiting for the noise to die out, so we can tell newbies they have toxic links'
* @Luzie (or someone?) deleted my new thread 'summary' of this problem 2 times; my original thread is bloated, trolled and lost.
Google put us in a room with you lot, and then ignored the room!
One guy, answering questions 'here and there' - there's nowhere else to get this issue raised. This is possibly the most ridiculous situation I've ever found myself in. I'm not going to hijack Dani's thread anymore, I'll post any follow up somewhere else.
g'luck to you all.
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/10/13 1:56 PM|
I can't "delete" anything, so I don't know what you're talking about.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/10/13 2:16 PM|
>>* @Luzie (or someone?) deleted my new thread 'summary' of this problem 2 times; my original thread is bloated, trolled and lost.
They were marked as duplicate. Don't shoot the messenger. I'm merely pointing it out.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/10/13 2:25 PM|
> So what would you do if a Googler suddenly chimed in and said "Yes, I confirm we've rolled out the Ghost Update in November 2011" - would that help you in any way?
Yes, it would help me. I feel like I've already been doing the absolute best job that I could possibly do. So I'm going to keep on doing more of the same. However, at the urging of someone else whose story was very similar to mine, it made me curious how many other sites were hit by the same thing and if this was a UGC specific issue and if it was actually something that Google meant to do or if we were again casualties of war, as DaniWeb was confirmed to be from an earlier Panda version.
Should a Googler confirm one way or the other, then it would simply give me the satisfaction of knowing the answer to the question I posed as my reason for starting this thread in the first place.
I actually take a bit of offense to the fact that Ashley marked luzie's post as the best answer, and subsequently the thread got marked Answered, because I have still not received the answer to the question I posed in my original post of this thread. As mentioned multiple times, my goal for starting this thread was to get a discussion going to see how many other sites were also affected the same way I was, and whether or not this was intentional or not on Google's part. I still don't know the answer to that question, and therefore I don't see why my question has been marked answered.
I am using the term penalty loosely in that Panda, for example, is essentially a penalty in that it's an algorithm tweak which doesn't work by raising the internal score of sites, but rather exclusively by issuing a sitewide hit (aka "penalty") to sites that fit its criteria. Because my hit was sitewide and not on a page-by-page basis, and the change was dramatic and overnight, I am considering it an algorithmic penalty.
Hitchhiker is referring to the fact that quite a few of the posts in this thread have subsequently been deleted by a moderator, and he believes that they were his posts and you deleted them.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/10/13 2:30 PM|
>>#1) We're now on a proprietary, unique platform now, and perhaps googlebot just doesn't understand / recognize it as a UGC forum because it's not like any other forum system they explicitly know about such as vBulletin, etc.
This is where things get confusing: on the one hand, you're saying that you got hit with some kind of update that hit UGC sites. But now you're saying the bots might not think of your site as a UGC.
I'm not pointing this out as a "gotcha". It just shows that you can drive yourself crazy with all the speculation.
But...it does bring up an interesting question: how does Google classify as site as UGC? Does it even do that? What if just looks at all the content and all the signals it uses to rank a site? How do the UGC people posting here know it was UGC related with all the billions of sites out there? Who's to say other types of sites didn't take a hit on that day and you don't know because you're not in the industry?
Just thinking out loud...
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/10/13 2:47 PM|
Daniweb caters mostly to places like India, etc. If you hit her forum, most of the content is not well written. I usually find what I'm looking for at Stack Overflow and don't have to deal with ads or popups or trying to figure out poor English. I'm sure she does well in India.
Google isn't here to give you traffic. Google wants to show whatever users want.
The algorithm isn't fair. It doesn't care about your money woes or your hard work or that you should get some traffic just cuz you worked really hard. It's an algorithm that detects what users want through data collection. It's ones and zeros. I was a daniweb user and she didn't listen to me, so it's showing in her results.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/10/13 2:56 PM|
>> Hitchhiker is referring to the fact that quite a few of the posts in this thread have subsequently been deleted by a moderator, and he believes that they were his posts and you deleted them.
We can't delete posts. Only a Google employee can delete posts, so he needs to point the tinfoil sombrero in another location.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/10/13 6:51 PM|
^^^^ Well stated. I didn't want to appear as if I'm hijacking a thread, just confirming Dani's and Frank's frustrations. I do much appreciate may of the regulars here who provide as much as they can - but they too are still outside of the holy temple. Sometimes we just need a little guidance from the all might, lol. :)
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||6/10/13 6:54 PM|
PS - and for all this talk about how some of our forums are dated, content not as valuable as those with a technology refresh. I'm still unable to do something as simple as edit my own post (like I could do in the 1990s) so that I can correct inadvertent spelling errors. Now if we really want to talk about what this does to the inherent quality of posts and the ability fo them to be properly indexed. ;) lol....
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/10/13 7:25 PM|
You can edit posts after a certain level I think.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/10/13 8:23 PM|
I actually take a bit of offense to the fact that Ashley marked luzie's post as the best answer, and subsequently the thread got marked Answered, because I have still not received the answer to the question I posed in my original post of this thread.
There was only a single question in your original thread (buried by all the "don't hate" disclaimers of things you obviously know are already a liability):
Apparently there's a theory that's out that just a few days before Panda 22, there was an unrelated algorithm update that specifically targeted large forums and UGC sites. Maybe I can get a bit of clarification on that?
No. You likely won't get clarification. And it doesn't matter. Google tweaks their algorithms hundreds of times per year and I felt like Luzie's answer was sufficient and poignant to that end.
Also thanks to Suz for duplicating threads. There is really no need to have multiple threads on this topic just because you don't like where the conversation went.
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/10/13 9:14 PM|
Ashley - do you ever post anything positive in these forums? You are so contentious. Everywhere I see you respond you like to dish out negativity towards others. Has anybody ever posted a website in these forums that you thought was actually good? Would love to see a positive side of you for once.
I also disagree with what your marked in this thread as answering the question. Nothing has been answered.
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/10/13 11:22 PM|
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||6/11/13 1:01 AM|
The original question concerned verifying an unrelated (panda) algorithm update that targeted a specific type of website and potentially a specific website. While not impossible for a Google employee to verify such information, in reply to a specific website issue, rather than a more general blog post or media question, the idea of using that information for some notion of search-engine optimization is probably not viable long-term.
There seems to be a focus on attempting to correlate some causation across multiple website statistics as a substitute for Google's validation. Again it seems as a notion of using that information to optimize a website against such as algorithm change. The rub of course is that Google tends to change the algorithm in response to the very same targeted search-engine optimizations which work against people-optimizations and people votes. What is changed on the website today could be offset by an unrelated algorithm change tomorrow.
The next few sentences will be boring, redundant, and will not verify any algorithm change based on paranormal activity, but may be seen as positive in the right LED light. Do no search-engine optimization. Don't fight against the algorithm. Make little if any changes based on the algorithm. Focus on amazing web sites that people will continue to like over time. Make changes based on people.
"It's hard to fight when the fight ain't fair." -- Taylor Swift
"Change means that what was before wasn't perfect. People want things to be better." -- Esther Dyson
|Re: DaniWeb hit in November '12 and haven't recovered since||Richard Hearne||6/11/13 2:02 AM|
Sorry Ashley, I like you personally a lot, but marking Luzie's answer as "Best" is wrong, plain and simple.
This place used to be about helping people. That rather simple purpose seems to have sadly gotten lost over time.
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||6/11/13 2:31 AM|
True love is telling it as it is. In the long run sympathy helps no one.
- not taylor swift ;)
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/11/13 2:35 AM|
It still is. Ask me about the configuration of your robots.txt, about redirects, site structure, whatever you want and you'll have an useful answer ... but please don't invent highly dramatic conspiration theories about "ghost updates" and expect me to give the comfortable answers that confirm your theories. Daniweb have been shown all the shortcomings of their site, possibilities for improvement have been thoroughly discussed in several lengthy threads now - all to no avail. The site is still a mess and daniweb in total denial as to what thin content in bad English, unwanted popups, deceptive ad placement and such do to usability and user satisfaction. They rather prefer to look at themselves as victims of "collateral damage" of "failed ghost updates", "Google bugs" and the like and are now trying to muster support for this kind of nonsense - I don't even understand what for - instead of taking care of their site. If help here consists in applause for mere Google bashing it won't come from me.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/11/13 3:01 AM|
It's great that people give advice, but don't be so definitive when giving it. If de-indexing pages, thickening up other pages, etc worked a lot more sites would have recovered from Panda, but as we all know very few sites have recovered.
If you haven't managed to get a site out of panda and know for a fact what works, your ideas are just guesses.
|Re: DaniWeb hit in November '12 and haven't recovered since||Richard Hearne||6/11/13 3:12 AM|
I appreciate a lot more than the average Joe that you guys are not equipped to answer the technical questions around algo changes.
But rather than dish site owners, and mark answers which don't address the questions asked I think honesty would be best - just tell people you're not able to conclusively answer, and leave it at that.
I think you and I agree that Google should not be sending webmasters who suffer at the hands of algo changes here at all.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/11/13 3:12 AM|
>>Also thanks to Suz for duplicating threads. There is really no need to have multiple threads on this topic just because you don't like where the conversation went.
It wasn't me. :-) I didn't see the original 'duplicate' posts that hitchhiker started, but if it was more of the same then duplicating the threads is a good idea if everybody is just going to repeat themselves.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||6/11/13 3:20 AM|
Has this thread really be marked as answered? I thought that was only done once a clear and definitive answer was given. I think her original questions was about a ghost update in Nov, did someone from google say yes or no?
|Re: DaniWeb hit in November '12 and haven't recovered since||SGT Oddball||6/11/13 3:53 AM|
This thread is clearly not answered and the answer that has been maliciously marked as correct is written by somebody that clearly didn't read the OP.
This is my first experience with google webmaster forums and I cant believe the hate filled rubbish coming from top contributor hobbyists, that have nothing to do all day but pretend they are experts, some very serious moderation is needed.
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/11/13 4:15 AM|
There hasn't even really been posed a serious question, I cite:
... aside from hearing me out, letting me vent, and
possibly getting the attention of the powers that be.
As to whether there was an update or not, the OP and her supporters are completely convinced of it, they don't even want to think of any other possibility, they have explicitly asked us not to discuss anything related to their site, so why do they need to have an update confirmed they know of as a fact? Now that this thread is evolving they suddenly realize it doesn't help them to have identified this update, it doesn't even help them to have a dozen supporters parroting them, their situation hasn't changed a bit.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||6/11/13 4:16 AM|
>>I think her original questions was about a ghost update in Nov, did someone from google say yes or no?
Putting aside the marked answer issue, does anybody here really expect Google to answer this? I'm not saying it's not possible, but the chances are pretty slim. They don't acknowledge many updates -- and I would think they don't even acknowledge tweaks to the algorithm.
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||6/11/13 4:28 AM|
Well ... to reiterate known facts, Google tweak "the algo" around 500 times a year, so chances are there was an update just on the date in question - and now what? Not that I had the authority to, but why not, I hereby confirm an update on November xth 2012. It hit daniweb and a myriad of other low to medium value UGC sites. Now what? The OP can mark any other answer as "best" now and we're done, everybody can go home and relax. Case solved :-)
|Re: DaniWeb hit in November '12 and haven't recovered since||Richard Hearne||6/11/13 4:35 AM|
There is always a possibility that some algo tweak caused unforeseen and unwanted changes. It's happened before many times, and on occasion been documented.
It's also likely that Google will fix issues quietly and say nothing. Tough-titty to any site affected.
I think it's perfectly valid for the OP to start this thread, but sadly some of the responses reinforce the fact that this Support Group is simply a way for Google to keep webmasters one step away from their staff. There is no "support" for far too many issues. And as Google increases the FUD and becomes less transparent with each passing day the problem will only get worse.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||6/11/13 5:37 AM|
Starting a thread in hopes that Google will use the information to tweak the algorithms at a later date or to garner suggestions to help improve a site is one thing and seems like a good use of the forum. Starting a thread to get a definitive answer about a specific algorithm update to a specific type of website or to help change the exact set of things that the algorithm affected is another and seems unrealistic.
There are best-guesses and suggestions and general feedback that might spark an idea and there are definitive change-exactly-this-and-you'll-get-back-what-you-had-before and yes-we-made-that-exact-change-on-that-exact-date and that-is-exactly-what-caused-the-statistics-you-are-seeing-on-those-dates.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/11/13 7:17 AM|
I just want to again point out that the reason why I kept saying don't talk about this, that and the other thing in the OP is because I already have other threads on this forum where I go into my SEO corrections in great detail. Been there done that. I don't need any more suggestions and feedback about what to change on DaniWeb because I'm personally content with my own site and that's all that matters. The purpose of this thread was meant to be a higher level discussion that was not DaniWeb specific.
I was approached on my own site by someone who claimed to have had a very similar experience to mine, and that prompted me to want to open up an intelligent discussion here in which I opened up my own Google Analytics data to the world and and asked if anyone else had a similar experience. My hope was to see if other people were affected and, if so, that they could compare their data to mine and see if, as a collective of people affected by the same algo update, we could come up with a more definitive idea of what might have caused it, what (if anything) we need to fix, and perhaps get the attention of TPTB and see if this was an intentional update for us or if we were just collateral damage.
For example, back in February 2011, people started posting their analytics data and saying that they were affected by some type of penalty out of no where but didn't know what. Once the first group of people posted something, more and more people started chiming in saying 'Me too!'. Together, the community put their heads together and was able to figure out who the big movers and shakers of the algorithm were, and consequently determined that it seemed to target content farms, and therefore dubbed it the Farmer update. A month or so later, upon lots of pressure to Google and lots of constant talk about what was going on, a Google representative finally came forward and acknowledged the update and said that internally Google was calling it Panda, named after one of the engineers behind the project. Over time, in depth analysis and sharing data amongst top SEOs to determine what some of the signals was helping people recover from it and figure out what google was targeting in the first place.
A few months later, Matt Cutts acknowledged in a YouTube video (link in one of my earlier posts in this thread, but here it is again http://www.youtube.com/watch?feature=player_detailpage&v=R7Yv6DzHBvE#t=1113s) that the way that google improves the algorithm is they read all of the feedback about it, and (to quote him), "We’ve made many, many changes over the last few months, even within Panda, trying to iterate, find new signals. You see a site like DaniWeb complain and then we find signals and say, okay, here’s a way we can differentiate between this site and the sites that might be a little bit lower quality."
That's all that I was trying to do here. I was hit by something that seemed to not fit any existing patterns of any existing algorithms out there. My first instinct was to say, "Hey out there in SEO land! Here's a detailed report of EXACTLY what I'm seeing. Is anyone else seeing the same pattern? If so, maybe we can put our heads together and come to the bottom of what our sites have in common that other sites that weren't hit don't." If you weren't hit by this algorithm, there was really no purpose to you contributing to this thread in the first place because there's nothing that you can contribute to helping the people who all WERE hit put their heads together.
My second goal for starting this thread was to get the attention of TPTB in much the same was as when Farmer was first being discussed. People put their heads together and had intelligent discussions about what might be going on, and then Google stepped forward and acknowledged what the community had come up with. Also, I'm coming from a place where Matt Cutts all but said that when a new algorithm is released, Google appreciates feedback as to which sites were hit, because, as with Panda, it's possible that there is collateral damage, and Google can then use that feedback to better tweak the algorithm into the future.
So what is so wrong that? If this question doesn't apply to you, then that's fine. If you're a top poster here, move on to the next question in the forums. But I don't appreciate being bashed for asking in the first place and then having my question marked as answered when it wasn't.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/11/13 7:23 AM|
Nothing wrong with that, but if you wanted roses and rainbows and OMG RANK DANIWEB from people who have watched for a while, you are in the wrong forum.
Gotta take the "I hate the popup" along with the tinfoilers agreeing with you.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/11/13 7:33 AM|
The reason I wrote in my OP to not bring up the popup again is because I debated the pros and cons of having it to death in another thread in this forum, and I wanted to have this discussion not go off tangent right from the beginning reiterating everything that was previously said elsewhere. I also didn't want the thread to turn into an OMG RANK DANIWEB or, alternatively, OMG DANIWEB SUX BECAUSE THERE'S A POPUP thread.
While it's fine that I appreciate you don't like the popup, we already spoke about that in my thread about recovering from Panda, and I subsequently recovered, popup and all. This thread was meant to be about an entirely different issue, and I was not intending for this thread to be DaniWeb-specific but instead open up an intelligent discussion amongst other people also affected by the same thing, and to also pull people out of the woodwork who are seeing the same thing and didn't realize that they weren't alone until they read this. Unless someone steps forward saying they were also hit and also have a popup, I'll consider the popup a moot point with regards to what's going on here.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/11/13 7:38 AM|
Too bad. I can bring up the popup as a reason why you lost rank. Just like it's within the ranters' rights to rant about Google targeting UGC sites unfairly, I can be more factual and tell you that the layout and the popup and the ads made *me* bounce, cuz I was one of your users.
At least I am giving you a factual reason why you lost me as a user compared to the tinfoilers.
My anatomy professor used to say something that sticks with me: If you choose someone to fix a problem (in this case medical, but it applies everywhere), you don't want the person who just sits complacent by your side boohooing over your problem and trying to make you feel better, you want the sick, egotistical mother f'er who doesn't give a shit about you but wants to figure it out because he is determined not to let it beat him.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/11/13 7:45 AM|
- Marking a thread "answered" certainly doesn't stop anyone from posting and the thread to be continuously bumped up for discussion.
- You can have multiple answers - pick one that you like best, if you wish! Answers mean nothing except to pin that post at the top so others don't have to wade through these incredibly verbose posts.
- I still stand by Luzie's answer - because we cannot give specifics about the algorithm and even if we could, it doesn't matter.
Richard - I know we're pals and we see things differently, but here's what I see:
-- Dani wants an answer to a question that none of us can answer. It's fruitless. The aglorithm is opaque. I'm sure you already found this page, right Dani? http://www.google.com/insidesearch/howsearchworks/algorithms.html it's got the best data that any of us can hope to ask for from Google on algorithm updates and it cited 665 updates last year. If you are looking for details on specific algorithm updates then this forum will always let you down.
-- Dani's site has had some issues in the past. She's received so much attention and help - but some things she wants to keep. That's fine. But saying what we can and cannot talk about isn't really helpful or reasonable.
I think the best solution would be to, if you want to discuss just algorithm updates (not tied to a specific site, but in general), start a thread in Chit Chat and hypothesize as much as you want OR post your URL and ask for general guidance. It's a broad net - but that IS how this place works and how it offers value. You're getting technical/strategic support from volunteers. It's unreasonable to think that we can deliver specifics on the algorithm or special circumstances because you felt you were targeted unjustly.
If you want to reach out to a Googler directly and cut this place out of the loop - I think that's reasonable. I'm sure you know a few of them on G+.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||6/11/13 8:59 AM|
> Too bad. I can bring up the popup as a reason why you lost rank. Just like it's within the ranters' rights to rant about Google targeting UGC sites unfairly, I can be more factual and tell you that the layout and the popup and the ads made *me* bounce, cuz I was one of your users.
That's perfectly fine if you want to keep the thread on track and, just like you just did, say that you think the algorithm hit might be because of the popup and might be influenced by bounce rate. That's fine: appreciate the feedback. But earlier when you just said that DaniWeb sux because we have a popup, and left it at that, it just makes me think to ignore you because you're just a hater and aren't offering any valuable constructive criticism at all.
But it does mark the thread as 'Answered' which might encourage others to not click into it.
There are no posts here that sufficiently answer the question I posed for me to be able to mark one as answered. I also don't see the benefit to the community in having to scroll past a 'Best Answer' that is disliked more than it is liked, according to the voting system here, not to mention that it doesn't even answer my question.
If you admit that none of you can answer my question, then why mark it as having been answered? On DaniWeb, we have a Computer Science forum, and on occasion someone may ask a question directed towards fellow computer science students. Such questions often don't have single answers, but are meant to get a discussion going about something or another. No one on the moderation team, myself included, is in a position to answer them because we are not the people the question was posed to. It's the same thing here: The question I posed *can* be answered: it can be answered by other people who were hit by the same thing that I was putting our heads together and having a rational discussion about it, and it can be answered by Google employees. Just because YOU are not in a position to answer my question (because you're not the demographic my question was directed towards) doesn't mean that you should bash me, mark it as Answered, and call it a day.
On my own forum, we encourage people to not start multiple threads asking the same thing, because it leads to too much duplication in responses, and we ultimately have multiple threads going about the same thing. All I was trying to do was say that I already have another thread here about DaniWeb, specifically, and I meant this thread to not be DaniWeb-specific. It is senseless to me to have two threads going both giving feedback specifically because I have not changed anything since the updates made as a result of last receiving help here.
The Chit-Chat section is an un-moderated amalgamation of non-English threads, blatant spam, and gibberish. No one with an intelligent question posts in there, and it's borderline insulting that you think that, after the amount of time, energy and effort I put into posting here, you think it belongs amongst those threads. I understand that you refuse to validate my question because YOU are unable to answer it, but I DO have a question and it IS related to crawling, indexing and ranking.
Contrary to what you might believe, you (and the other regulars here) are not the only ones who are allowed to answer questions here. My question was directed towards the community at large to see if anyone else is facing the same issue that I am, and what it is that all of the affected sites have in common. We figured out, through the progress of this thread, that we are all UGC, but the question still stands what sets us apart from all of the other unaffected UGC sites out there? What is it about our "quality" that Google feels isn't as good? There has to be a pattern and that is what I'm hoping to find the answer to.
If I understand correctly, you have the ability to escalate this thread to Googlers. If you think it's reasonable for a Googler to be directly contacted regarding this issue, then by all means. But that's your call, not mine. I'm just here because I saw something strange on my own site, and thought it was the most appropriate place to reach out to the community at large and gain feedback regarding the scope of what has affected me.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||6/11/13 9:20 AM|
You're a peach, Dani.
Good luck with the site.
The thread was escalated a week ago.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/11/13 9:25 AM|
Your sense of entitlement is showing, Dani.
|Re: DaniWeb hit in November '12 and haven't recovered since||SGT Oddball||6/11/13 9:36 AM|
Your sense of being a troll has shown since your first post in this thread.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/11/13 9:37 AM|
annnnnd it's sock puppet time!!!!
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/11/13 9:47 AM|
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||6/11/13 10:18 AM|
Requesting feedback only from similar website owners who had similar statistical trends to confirm an algorithm change or a confirmation directly from Google, was implied but could have been more direct.
But why would you want this information? To what end?
There are probably dozens of user-driven changes that could still be made to improve the site without any knowledge of Google's internal changes.
OK then. Sounds like reacting to the algorithm is the focus.
The algorithm, and the algorithm, and the algorithm,
Creeps in this petty pace from day to day,
To the last syllable of recorded time;
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||6/11/13 10:20 AM|
Lysis, this has nothing to do with the Ghost update, none of the other UGC sites that I have seen have this. So this is not helping answer any questions here.
This thread isn't about critiquing Dani's website, she already mentioned she has been getting help in her other thread about what should be fixed and have taken most of the suggestions there seriously. This thread is about discussing what may have happened in mid November to UGC sites -- and while Dani's site appears to be one that was hit, there are a bunch of others that were as well. The main common factor that we can all figure out so far is that we are all UGC sites, some of us rather large. The other common factor is that many of us care a great deal about our members, and aren't trying to rank for the short term. Most of us don't follow keywords, we focus on our members and creating a useful website.
While personal preferences vary between websites, I have learned very well that there is no way you can please everyone. No matter what you do there will always be a few that don't like your site, and there will be a ton who love it and appreciate everything you are doing. I feel that many of the UGC sites hit by this update fit in that boat, they are all created by experienced webmasters, programmers, etc who are working to try to better the internet. These are not people trying to manipulate the system, but just trying to understand what happened in mid November and if it was an intentional update geared towards UGC sites, or if UGC sites were collateral damage in some sort of other update.
Unless you are a Google employee, or have a UGC site with some information on what is happening to you, the rest of you are wasting everyone's times because you have really no idea what you are talking about.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||6/11/13 10:28 AM|
>> Unless you are a Google employee, or have a UGC site with some information on what is happening to you, the rest of you are wasting everyone's times because you have really no idea what you are talking about.
This internet forum
Internets forum open for replies.
You add to replies.
You kettle, me pot. You black too, kettle.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/17/13 5:56 PM|
I've taken some time away from posting in this thread because it was simply turning into a flame war and that's not my style.
I've gone ahead and just posted a news article on DaniWeb explaining my situation, which I'd like to point you guys to:
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||7/17/13 6:17 PM|
But if stackoverflow stole your traffic , then they also showed you the way to rank again. The answer is right there.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||7/17/13 6:22 PM|
For the first example Google shows DaniWeb as the first five results for me, albeit collapsed. So the conspiracy related theories do not seem to hold a lot of water.
I've used DaniWeb a lot in the past but when there's a choice between navigating an ad mine verses a better user experience overall, the choice is rather easy unless there is some other overarching factor attracting my attention.
I knew nothing of how either site was funded when making my choices. The choices were based purely on user experience and then bookmarking and returning to the site. Google's influence was rather secondary since I migrated to stackoverflow before the results reflected my preferences. But surprisingly, I suppose, Google did eventually reflect my personal preferences even when on public computers.
"Public opinion in this country is everything." -- Abraham Lincoln
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/17/13 6:50 PM|
> For the first example Google shows DaniWeb as the first five results for me, albeit collapsed.
Did you disable personal results?
I'm not saying that StackOverflow shouldn't rank well. I'm simply saying that it seems to be ranking well NO MATTER WHAT, even in cases where they are simply doing little more than linking out to other sites that have the complete answer.
I'm also saying that while they definitely deserve to rank well, they don't deserve to be the ONLY site that ranks well. Owning nine out of ten results for all discussion-based results with the broad keyword "c++" is a bit too extreme IMHO.
Google should be spreading the wealth around a bit more. Not *everyone* appreciates the Stack Overflow format, and Google is creating a monopoly here. IMHO no single broad keyword should ever be nearly completely owned by a single domain unless the keyword is a brand.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/17/13 6:55 PM|
Also, the point of my article wasn't to point out that we lost traffic, but simply to demonstrate that at the exact same moment Google started sending less traffic to us in favor of StackOverflow, the amount of referral traffic we received from StackOverflow increased by a proportional percentage.
My point is that I have proof that Google is sending its users on a wild goose chase because instead of just sending visitors directly to the answer to their query, they're instead sending visitors to a middleman site that is redirecting visitors to where the answer is.
Not in all circumstances, because StackOverflow is still an amazing site and deserves to rank in many cases. But the fact that they are sending traffic to StackOverflow even in cases where the StackOverflow landing page doesn't contain the answer directly is proof that the algorithm is too biased towards the stackoverflow.com domain as a whole.
|Re: DaniWeb hit in November '12 and haven't recovered since||redleg-redleg||7/17/13 7:02 PM|
-> I'm not saying that StackOverflow shouldn't rank well.
I'll say it StackOverflow should not rank that well. It comes up, 3-4 results, on 90% of my searches and I have yet to see a useful/helpful answer. Most of the time I do not bother to look at any of the hits but when I do I can not begin to understand why the page was even returned.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||7/18/13 1:36 AM|
It looks like stackoverflow.com has lost a lot of traffic recently.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||7/18/13 3:18 AM|
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||7/18/13 5:15 AM|
But why do you say StackOverflow "stole" your content? That title made it difficult for me to take the rest of the post seriously. StackOverflow itself didn't "do" anything to daniweb.com
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||7/18/13 6:17 AM|
lol @ opening the link to get an annoying popup in my face.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/18/13 7:49 AM|
> But why do you say StackOverflow "stole" your content?
Because, like all mediocre news and gossip writers, I had to come up with some weird title that's catchy and yet never seems to 100% accurately describe what's really going on, just to hook people in.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/18/13 10:29 PM|
Oh, and plus because I meant it in a very tongue-in-cheek way, which is obvious upon reading the article. Maybe it's just because I'm a New Yorker??
|Re: DaniWeb hit in November '12 and haven't recovered since||ets||7/18/13 11:21 PM|
Then I'm afraid it was a complete "fail" for me, because I took one look at the title and didn't go any further - and I have some sympathy with your arguments (having followed since 2011). I have looked at it now. IMHO, the "spread the wealth around" argument has a lot of merit; on another thread, you can see several people complaining about Yelp. But I think an argument like that has to be made by referencing a whole range of sites - not just your own - in a much more objective way, or it simply isn't persuasive; admittedly, you don't have the detailed statistics for other sites that you have for your own. But that's how I'd frame that argument - and probably all your arguments. They're failing to convince because they're all about your site. If the same arguments had been made by Joe Objective, ace reporter-at-large for Search Engine Land, citing evidence from a range of different sites, I think people would have much more time for them.Oh, and plus because I meant it in a very tongue-in-cheek way, which is obvious upon reading the article.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||7/19/13 8:55 AM|
Half the time I find it dizzying just trying to read StackOverflow and appreciate the front facing interface.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/19/13 12:49 PM|
> But I think an argument like that has to be made by referencing a whole range of sites - not just your own - in a much more objective way, or it simply isn't persuasive; admittedly, you don't have the detailed statistics for other sites that you have for your own.
Exactly. I can give some anecdotes about other sites in the same exact predicament, such as DreamInCode, but I don't have the stats to back it up. The point of my article was simply to showcase the inverse proportional graph between our own decrease in Google traffic and increase in referrals from StackOverflow.
I would have no way of being able to gather such referral statistics from others.
Essentially I started the article to say "Here's what I'm witnessing. I've heard stories that other people are in a similar situation. For those of you out there who are, care to share ... ?"
Unfortunately, I've found very few people are willing to put their statistics out there the way that I have always been willing to do.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||7/19/13 3:01 PM|
I honestly don't know what to make of this thread. I think the statement speaks for itself and I have to ask... why?
Moving forward, we are officially closing this Google Group.
Use the Google+ Community to discuss the platform, get feedback on your ideas, and meet new people with similar development interests.
If you have specific technical questions, post on StackOverflow using the tag ‘google-plus.’ The Google+ Developer Relations team monitors this tag daily to address your questions and help you develop with the Google+ Platform.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/19/13 3:35 PM|
Wow ... just wow.
The statement that Google is publicly closing their support forums in favor of redirecting its users to StackOverflow expresses in no uncertain terms that there is favoritism there.
Fine, let there be favoritism, but PLEASE don't create a monopoly by allowing that favoritism to expand so much in the SERPS as to allow StackOverflow to put all other programming communities out of business.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||7/19/13 5:50 PM|
That's old news. It's been that way for almost 2 years.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/19/13 7:28 PM|
Well I don't doubt that Google hasn't been pushing and showing favoritism towards SO for a few years, but that was just posted a few weeks ago.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/19/13 7:31 PM|
Oh ... and before I get heat for this, I just want to clarify. When I say that I am confident here that Google is pushing and showing favoritism towards SO, I mean in this type of context where they are clearly vocal (as a company, not a search engine) about redirecting their end-users and customers towards StackOverflow.
I make no accusations (other than unfounded conspiracy theories) that they have manually tweaked their algorithm to favor SO.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/19/13 9:11 PM|
Another issue I found. It appears that my site is mis-classified as all of its pages fall under the Google filter 'Blogs' instead of 'Discussions'.
Is there someone who can escalate this to a Google employee to correct or is it something I can correct on my end? I'm already using microformats for breadcrumbs and to indicate the authors of each post in the threads.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||7/21/13 8:13 PM|
don't we usually favor the ones we like best?
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||7/21/13 10:26 PM|
I'm not disagreeing with that.
I'm simply pointing out that it's not necessarily serving Google's users' best interests to have eight out of ten results in the SERPS all be to different pages within the same domain. It doesn't matter what the niche is or what the domain is. IMHO, the purpose of a search engine is to provide the user with different options of where to find information on the web given a particular search query. Having all of the results point to the same website is not as useful as having a mix of different websites for the searcher to peruse.
Additionally, favoritism is one thing, but I think they have a bit too much of a sitewide boost when SO pages are outranking DaniWeb pages and the only meaningful content on the SO page is a link to the DaniWeb page.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||7/22/13 1:37 AM|
I feel your pain, in my career I've had two sites which have been hit by google. In both cases I don't think the penalty / downgrade was fair, and it certainly wasn't fair when you look at the crap and rule breakers that google does decide to rank instead of my sites.
I've tried and tried to fix both sites, but with no luck, it feels like once hit you have a permanent black mark against your site. So this only leaves three choice:
Google has a very thick skin, after ever major update there are a lot of casualties and collateral damage, who try to get google to put things back to how they were but this doesn't happen. Every year their adwords revenue goes up as do their profits, and this is what counts for a public company.
Your arguments are valid but the situation won't change, so wouldn't you better off if you change tack?
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||7/29/13 10:52 AM|
Guys you keep on mentioning S.O, although different in 'timing' you'll notice they experienced the huge drop too:
They may well have been manually corrected to compensate (during that 'soften' update) last week.
Details, who knows!? - but it's at least interesting. If S.O got hit so hard, what does that say about the algo?
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||7/29/13 10:58 AM|
Not exactly sure why Alexa is so wrong there, seems you gotta take Alexa with a grain of salt sometimes. Fortunately stackoverflow has its stats available to the public via Quantcast:
You will see they were not affected at all this year, they have only improved. Maybe a very slightly slide bringing them down to around 3 million people per day.
|Re: DaniWeb hit in November '12 and haven't recovered since||luzie||7/29/13 11:02 AM|
hitchhiker, I don't see that huge drop, at least not on the graph shown on Alexa?
There's something else remarkable in this report:
How engaged are visitors to stackoverflow.com?
Daily Pageviews per Visitor
Daily Time on Site
Daniweb should compare their own figures to that ... stackflow show good values for the metrics involved, a signal of user satisfaction that translates into good rankings.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||7/30/13 3:29 PM|
Basically they are measuring different things. Alexa's graph shows the 'order' rather than volume. Whatever happened, the level at which SO operates on means that 'statistically' the error rate up there is low. Something did happen (unless it was an Alexa bug).
When I used to measure with quantcast (i had around 180k a day) - I noticed it only seemed to measure certain parts of my audience. The Alexa pattern *seemed* more 'evened' out, following the private stats quite accurately at volume. (Alexa is useless below 50k a day, as the number of toolbar users is relatively low.)
Who knows, hell of a coincidence though?
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashiq Hassan||7/30/13 5:05 PM|
why you people are not answering my question
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||8/21/13 2:57 PM|
Pardon my absence, I was off creating a new skin for DaniWeb. Not going to sit back and just bitch and complain without doing something about it :)
Regardless, here's where my frustration lies. I have *NO ISSUES* at all with Google ranking StackOverflow #1 in the search results. The problem is that Google is creating a monopoly out of SO by giving them positions #1, 2, 3, 4, 5 ... and effectively forcing out every other player in the niche.
In doing my redesign, I was legitimately doing a google search for "css bottom toolbar" because I wanted to get some design ideas to redesign my bottom toolbar. I ended up being a frustrated Google end-user because all of the results on the page were ONLY to stackoverflow.com!
I'm of the belief that the purpose of a search engine is to take the searcher's query and then present the searcher with a multitude of different options of where to find information about that search query on the web. It's frustrating for every page in the SERPS to be from the exact same domain considering it's a very generic query (i.e. the company's name isn't in the query).
This just keeps happening to me over and over again and THAT's why I'm frustrated!! Google has literally positioned StackOverflow to squeeze out the possibility of any other programming site ranking well in the SERPS.
As far as my niche is concerned, I would consider myself the #2 ranking site in terms of traffic behind StackOverflow. The difference is that SO has about 100X the traffic that we do. And it *ALL* comes from Google. I'm not complaining because #2 isn't a shabby place to be at all. But I think that there's something a little off balance when there's a 10,000% gap between the #1 ranking site and then "all the rest". It's sorta like the whole "we are the 99%" slogan.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||8/21/13 4:34 PM|
>>In doing my redesign, I was legitimately doing a google search for "css bottom toolbar" because I wanted to get some design ideas to redesign my bottom toolbar. I ended up being a frustrated Google end-user because all of the results on the page were ONLY to stackoverflow.com!
Maybe it's personalized results? I only see 4 from SO. Then I get namepros, phpfreaks, webmasterworld, kirupa and css-tricks.
I've come across your site while searching for code. Unfortunately, I skip over it now because I've accidentally scrolled over ads and had that lightbox effect pop up. Startles the crap out of me and breaks my concentration when in the middle of a coding problem.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||8/21/13 5:08 PM|
> I've come across your site while searching for code. Unfortunately, I skip over it now because I've accidentally scrolled over ads and had that lightbox effect pop up.
We switched to it only doing the effect if you keep the mouse hovered over the ad for a duration of time (a LOT less intrusive) awhile back, to not doing it at all on article pages anymore as of our new skin that rolled out last week. No more lightboxes or welcome modals of any kind anymore.
The same whether logged into google.com or not logged in, and with personalized results disabled. Cookies in browser recently cleared as well.
Stackoverflow.com is the first 7 results for me, and then #8 is namepros.com, #9 is phpfreaks.com, and #10 is webmasterworld.com, and that's all I see on the first page of the SERPS. Honestly though, this is just one example. This happens to me *all* the time, which is where my frustration lies. I really don't want to see more than 2 or 3 results from the same domain name on the same page, ever. And, preferably, I just want to see it rolled up into a single result with sublinks to other related / similar pages that match my query.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||8/21/13 5:20 PM|
> No more lightboxes or welcome modals of any kind anymore.
Just to clarify so I don't get any slack:
- No more welcome modal or modals of any kind
- Lightbox hover effect when hovering over the ads on our homepage and navigation pages, but only if you keep the mouse hovered over the ad for a duration of time, so it won't suddenly happen if you're naturally moving your mouse around the screen
- No lightbox hover effects on any content pages anymore at all
Regardless, *my* ranking isn't my frustration. My frustration lies in that StackOverflow is taking over ALL the spots in every search query in my niche, even for stackoverflow.com pages that do little more than just say "the answer is on daniweb.com". StackOverflow is currently my #1 referrer source outside of Google, even beating out Yahoo and Bing!!
How is it a good end-user experience for half the page to be SERPS results for the same domain name?
How is it a good end-user experience for a SERP result to be a page that just says "click here for the answer on a different domain" instead of pointing the end-user directly to the page with the answer on it right from the SERPs??
|Re: DaniWeb hit in November '12 and haven't recovered since||Ashley||8/21/13 9:28 PM|
If you have questions about search in general, use the Submit Feedback link in the bottom of the query or post in the websearch forum
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||8/22/13 5:30 PM|
Thank you for your suggestion. I repeated the search and clicked on the "Send feedback" link at the bottom of the page and typed the following description:
Lately I've been doing a lot of searches where 5+ of the results are all to stackoverflow.com ... In this particular instance, I was looking for some design ideas using only CSS for bottom toolbars, and 7 of the 10 results on the first page were to stackoverflow.com, a site which is useless to me in this instance because I already know *how* to do the CSS, I just want some ideas and suggestions from a UI perspective. I was hoping Google would offer me a broader range of options. It's very frustrating to have nearly the entire page of the SERPS take me to a single domain name. I want choices, please!!
I will post in the Google product forums when I get a chance.
It's very frustrating that Google is allowing StackOverflow to literally squeeze out every other site in the entire niche by giving them not just the #1 spot, but also the #2-10 spots.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ben Griffiths||8/23/13 5:54 AM|
Why would google assume [bottom css toolbar] was a design and not a coding question?
It seems your input is flawed, not the output.
What does the SERP for [bottom toolbar css ideas|design|examples] look like? All StEx?
|Re: DaniWeb hit in November '12 and haven't recovered since||True Answers No Hypocrisy||8/23/13 6:02 AM|
dani u don't have any chance unless google start fixing algos
number 1 : don't overrank mass sites
number 2 : show only 1 result per site
number 3 : don't show links from repeated sites, there are many sites exactly dulicate with different theme
number 4 : rank sites according to their authority and quality not based on how less they are penalized
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||8/23/13 8:16 AM|
> Why would google assume [bottom css toolbar] was a design and not a coding question?
I was expecting a combination of both. Regardless, my frustration wasn't that all the results were wrong and I needed to fine tune my query, but that 7 of the 10 results on the page, with personal results turned off, were from the same domain name. My frustration comes from this not being a one-off occurrence but rather stackoverflow owns more than half of the search results quite often for me.
"Bottom toolbar CSS ideas" gives me 4 StEx results with personal results enabled and only 1 with personal results disabled. Needless to say, *none* are ideas at all.
With both personal results enabled or disabled, "Bottom toolbar CSS examples" has StEx in the #1, #2, #3 and #4 spot.
Harder than you thought to come up with a query that doesn't include StackOverflow more than twice on the first page of the SERPS, eh??
|Re: DaniWeb hit in November '12 and haven't recovered since||Ben Griffiths||8/23/13 8:20 AM|
I didn't do the queries, I queried your search intent.
I do laugh at how often Stack Exchange comes up in results I myself use, FWIW.
Please don't attribute any built-in intention to defend Google to my comments.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||8/23/13 8:39 AM|
As Ashley stated, I'm not intending to derail a conversation about my site with a grudge about Google's algorithm in general. I'm simply offering up the option that the answer to why my ranking went down was not because of anything specific with DaniWeb, but rather that the ranking of *all* programming sites went down because Google is allowing Stack Overflow to push everyone else completely out of the SERPS. I backed my theory up with proof.
What I do find amusing, however, is that the less traffic we get from Google, the more traffic we get from referrals from Stack Overflow. In tracing the referrals, it's clear that Google is ranking Stack Overflow even on pages where all they do is link out to the other programming sites that got pushed out of Google.
Something's got to give here, because Google is creating a monopoly in the pogramming community niche, and it's not a pleasant experience for Google end-users.
|Re: Spam collector thread||True Answers No Hypocrisy||8/23/13 9:28 AM||<This message has been deleted.>|
|Re: Spam collector thread||True Answers No Hypocrisy||8/23/13 9:28 AM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/4/13 4:27 PM|
Sorry to ask this question a second time, but I don't believe I received any responses to it the first time I asked, a page or two back, and would be grateful if anyone could shed some light.
Does anyone know how to petition Google to reclassify most of my site as 'Discussions' instead of 'Blogs'?
Currently, if I do a site search, if I use the 'More' dropdown and filter our results by 'Blogs', all of our forum threads show up. But none of our pages are listed if we filter by 'Discussions'. We don't have any blogs on the site and we're pretty much entirely forum discussion threads.
Additionally, each post uses rel="author" to link to the poster's member profile, so Google should be able to tell that there are multiple posters per page. The first post in a thread indeed looks a bit different (which might give the impression that it's a main article and the replies are comments to it??) but it's entirely stylistic -- all of the HTML (with the exception of one or two css classes) is exactly the same, and even then it's not very far removed from forum-looking.
|Re: The SE_Expert/MasterOfPuppets Thread: Diary of an Angry Fail SEO||True Answers No Hypocrisy||9/4/13 6:42 PM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/4/13 7:52 PM|
Not quite my area of expertise and i don't run a forum, but for what it's worth... I checked out another forum to see if the same thing happened. The only one I could think of offhand was stackoverflow.com (sorry!). Their pages show up under Discussions -- but they don't use rel=author. It would seem to me that rel=author is for 'content' -- which I interpret as an article.
It's really late for me and I only skimmed this Google blog post on rel=author but it seems there should only be one author per page, from what I can see: http://googlewebmastercentral.blogspot.ca/2013/08/relauthor-frequently-asked-advanced.html Maybe that's why Google's interpreting your site as a blog instead of a discussion? Check out how other forums are doing it and whether they're showing up under Discussions or Blogs.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||9/5/13 3:09 AM|
If google is classifying daniweb as a blog due to the use of rel author there is no wonder why the rankings are poor. The content is extreme thin for a blog. But probably ok for a forum /q&a site.
Suzanneh might have just found the fundamental reason why your rankings are poor.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/5/13 7:34 AM|
> If google is classifying daniweb as a blog due to the use of rel author there is no wonder why the rankings are poor. The content is extreme thin for a blog. But probably ok for a forum /q&a site.
Agreed. I pointed this out a few pages back but no one replied about it, which is why I brought it up again. However, it never occurred to me that the reason why I am being classified as a blog is because of multiple rel="author" links. I was under the impression that was the correct way to do it.
I'm getting mixed opinions as to whether it's okay to have multiple rel="author" links per page. I noticed that Xenforo.com forum system uses the http://data-vocabulary.org/Person microdata next to each post, but I could have sworn that they were the ones who originally gave me the idea to use rel="author".
One of the biggest issues I'm facing is that the information out there seems to be ambiguous as to whether they are referring to rel="author" attached to individual links, or the rel="author" metatag in the HEAD of the document, which would of course refer to the document as a whole.
http://webmasters.stackexchange.com/questions/25140/how-to-implement-rel-author-on-a-page-with-multiple-authors => In favor of multiple per page
http://googlewebmastercentral.blogspot.com/2013/08/relauthor-frequently-asked-advanced.html => Against multiple per page
It's confusing because the above blog entry in Google Webmaster Central, they seemed to be against multiple rel="author" per page. However, the following is a quote from Matt Cutts:
I'm pretty excited about the ideas behind authorship. Basically if you can move from an anonymous web to a web where you have some notion of identity. And maybe even reputation of individual author. Then web spam, you kind of get the benefits for free. It's harder for the spammers to hide over here and in some anonymous corner... If Danny Sullivan writes something on a on a forum or something like that I'd like to know about that even if the forum itself doesn't have that much PageRank or something along those lines so... so I do expect us to continue exploring that. We might be able to help improve search quality... and I think will continue to look at it to see how to use rel=author in ways that can improve search experience.
Basically, Matt Cutts is using a post on a forum as an example of a best use case for rel="author".
I have to run to temple now (Jewish new year and all that) but I'll investigate this more when I get home later. If anyone else could shed some light while I'm gone that would be great.
|Re: DaniWeb hit in November '12 and haven't recovered since||Ben Griffiths||9/5/13 7:38 AM|
I'd go with Google... MC's statements seem very forward-facing so not something I'd use in short term decisions.
It's early days for rel=author, I'd be terrified to deploy it as you have given the scale and the possibly negative or confusing signals.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||9/5/13 7:52 AM|
Yes I agree with Ben, what matt cutts was saying is how rel-author could be used in the future.
Whereas the google webmaster blog is stating how rel-author should be used now.
Another consideration is how useful rel - author is on a site like daniweb where the users might not use their "real" identities that carry weight. I think you said that you have a million members my guess is that 99.9%+ are not famous / important authors like danny sullivan. So do you really want google to think that 99.9% of your content is written by authors which dont have authority elsewhere?
If stackoverflow is out ranking you all the time they must be doing something right, so best to adopt their approach because something they do is working.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/5/13 5:15 PM|
Removed rel="author" ... Let's see if we start showing up with any Discussion pages.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||9/6/13 1:26 AM|
We began getting our traffic back (2 weeks ago) - now up 30% to around 130k daily. We also don't show up under discussions, the blogs above us do though, we are using 'rel-author' correctly to identify our posters. I'm guessing I have to remove that now?
|Re: DaniWeb hit in November '12 and haven't recovered since||LarryDarry||9/6/13 2:12 AM|
Daniweb's site is ranked super high in Alexa in the 3k level. She must be a millionaire. At first I opened this thread thinking it was some poor smaller blogger suffering from a panda hit or something with only a handful of views a day and no ad revenue. Her concerns are legit since many people suffer from google changes but is this a monetary issue more than anything?
Sorry but this part made me laugh:
I gained 5 pounds since I started working from home
Are we supposed to feel sorry for you because you put on 5 pounds working at home? The OP sounds like a princess and is panicking because her high earnings have come down just a bit.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||9/6/13 2:26 AM|
a) You know nothing about being in Alexa 3k - don't make assumptions beyond your scope.
b) You are calling a poster a 'princess' - think about how redundant that is on a tech forum.
c) You (I assume) haven't read anything beyond the last post, your value is limited.
If you want to contribute something, read a bit, think about it - and get back to everyone.
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||9/6/13 2:35 AM|
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/6/13 3:48 AM|
Was that around the 23rd of August? There was some chatter about an update around then; nothing confirmed. Or do you usually get a boost in traffic when school starts?
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||9/6/13 7:47 PM|
On Friday, September 6, 2013 4:26:16 AM UTC-4, hitchhiker wrote:
I never understood why rel=author would apply to the forums as it seemed always to apply to blogs. The impression I received was that original articles of value appearing in blogs and publications would have recognized accreditation - a stamp of the author's brand recognition. It didn't seem appropriate to similarly elevate and highlight every first post someone made in a forum thread that might consist of 100 words venting about being refused the use of the rest room at a local Starbucks or a 20 word slam of a terrible film they had just seen. That said, I'm not so sure that I recall rel=author being made totally clear and left somewhat vague, hoping to see how far this concept could extend.
Some forum owners posed the question of whether it was a good idea to use them for forums, especially how much it could destroy the privacy factor that many enjoy. First adopters hoped to have more conspicuous results appear on SERPs so they went ahead and tried it. I wonder whether Google classified these forums as "blogs" since blogs seems to have been the spirit of what they were intended to do originally.
Speaking of rel=author, I didn't realize that Matt Cutts was spending so much time in India. Seems like a number of blogs are using his picture to game viewers, lol.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||9/6/13 8:12 PM|
On Friday, September 6, 2013 5:26:45 AM UTC-4, hitchhiker wrote:....
Perhaps it might be best if this thread ceased and Dani could create a new one with a new subject line so that it wouldn't lend the appearance of an ongoing campaign - problems solved. I think Dani had considered doing that at one point and perhaps it's the right time.
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||9/6/13 11:43 PM|
Was that around the 23rd of August? There was some chatter about an update around then; nothing confirmed.
Hi Suzanneh (I assume you were replying to me?) -> We don't get any boost around this time of year, it was the 16th.
In our experiences updates come (to us) a week before anybody else's.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/7/13 5:28 AM|
@hitchhiker -- yes that was addressed to you. :-) I did quote your post but sometimes the quote shows up as a little "show quoted text" link; not everybody notices it. Glad that you're seeing some improvements.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/7/13 4:07 PM|
LarryDarry, so you feel like a wealthy businessman who has a very lucrative, fulltime job, and has a small blog he does on the side as a hobby that got killed from Panda, is very deserving of help and support? But someone whose only source of income is their website, which, while a popular site, lost two-thirds of its traffic (and therefore two-thirds of the entire household income, to the point of no longer being able to pay their mortgage) is not deserving of help because their site is 'too big'?? I suggest that you judge websites on their own merits and not on preconceived notions of who you think runs them.
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/8/13 4:55 AM|
i was in a similar situation, decided to expand my business in terms of providing direct coding and linux support to companies ( + a 2 day job )... this helped enormously to not be dependent on websites anymore... especially if your traffic comes from google you are at risk... one day, your income seems to go through the roof ( seeing yourself as a potential millionaire )... and a year later, your income can be cut in half... i learned that the hard way... all sites income now go into a fund... its nice to not be dependent on google anymore. I advise all of the ppl who are dependent on google traffic to have at least one fallback... best to have 2. in this economy.
|Re: DaniWeb hit in November '12 and haven't recovered since||Marie Haynes||9/8/13 5:46 AM|
Personally, I don't think that Google is consciously trying to give StackOverflow more top positions than any other site. It's more likely that these pages are more helpful to users than others. If I was looking to figure out the CSS to put a toolbar at the bottom of my page then each of the 4 top SO answers that I see is actually quite helpful.
I do a lot of coding for my sites even though I have no training outside of what I have taught myself on the internet and I do a lot of searches to figure out how to code something. When I see a StackOverflow answer I usually click on it because that's usually where I end up finding my answer.
So, I think that the problem here doesn't lie with Google but rather with how you have searched. I would think that the majority of people who search for CSS bottom toolbar would be looking for information on how to code something like this. As such, the results that Google provided are good ones. Now, if I searched for "design examples css bottom toolbar" I get results that are closer to what you were probably looking for:
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||9/8/13 6:37 AM|
On Sunday, September 8, 2013 8:46:02 AM UTC-4, Marie Haynes wrote:
The treatment is not specific to Daniweb. I've noticed over the past year that Google will provide a disproportionate number of results for the same site. If you search in the sports niche, I've found Bleacher Report frequently appearing with numerous results. I wouldn't say any of the additional articles were any more helpful than another.
I also wonder about Google's personalized results and whether your visiting a site once or perhaps just a few times may create a domino effect against competing sites appearing in personalized SERPs.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/8/13 8:13 AM|
> I also wonder about Google's personalized results and whether your visiting a site once or perhaps just a few times may create a domino effect against competing sites appearing in personalized SERPs.
Personal results turned off.
|DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 10:20 AM|
I think a pet talk is now appropriate (what do you think?)
DaniWeb, my desperate sweet one, honey of the hive, (the hive you no longer have) well so you say! the same one that became fat (to yourself), the same one that has not yet been answered (that part I find a little crappy)
Sorry to say, but, I can't help but laugh at what you just said.
You can't turn it off!! Its their excuse for the fu@ed results.
How on earth can you switch it off if it don't work!!?
And actually (believe it or not, eventually) I came across a "google employee" (by the way thanks for forwarding me to a link) that did stipulate that noindex does indeed follow links, but they also stipulated that a noindex means exactly that, nothing!! they very seldom come back to re-crawl that page. Think about it.
And they also mentioned (which I have always stated) you spamming. Yes indeed you are. You spamming the title tag. Even thought (some think) they don't recognize it (for the specific reason it don't come up with the text version) They do indeed take it as a spam indicator, and you stuffing yours on all pages with keywords!!
Take note what John said in one of his comment. "It looks like a lot of your content is hidden behind title attributes. To our algorithm that might look like spam. To our algorithm, that might look a bit sneaky -- "
And yours are spam and throughout your site. In other words "excessive".
Sorry about no links and all that crap, I'm typing on my blackberry (the same one that views your site very crappy because of your titles overpowering my small screen!!)
Think about it, that's all.
I've called it a spam thought for the day!! and don't take me up wrong, this ain't another "I hate Dani" speech, you filtered because of spam and the way you blocking the crawlers.
Hey, that's me. I post crap!! Eat it or regurgitate it.
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 11:02 AM|
Ok, i already saw the one downer no click (small change i say, i get plenty of them)
My point is, do you actually think crap liker this will work?
"title="Our Internet Marketing forum category encompasses everything related to running, managing, maintaining, and growing a successful online presence. To enjoy candid discussions about business practices and strategy, please keep the specifics out of the discussion so that members can feel more comfortable without the fear that their competitors, potential clients, or website visitors will know who they are. To keep everything anonymous and encourage open, unbiased and incentive-free discussion, please do not refer to your site by name nor link to it anywhere within the Internet Marketing category. For feedback and constructive criticism on your website, please visit our Website Reviews forum within the Business Exchange category. To do business with other webmasters in our community, please visit our Webmaster Marketplace, also within the Business Exchange category."
When last did you look at all your spam?
Your home page (and entire site) is keyword stuffed!!
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 11:10 AM|
Your stairway to "heaven" will never be glitter and gold tell you drop all the spam
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 11:40 AM|
Now dont you think this is pushing things to the extreme?
<a href="/internet-marketing/25" title="Our Internet Marketing forum category encompasses everything related to running, managing, maintaining, and growing a successful online presence. To enjoy candid discussions about business practices and strategy, please keep the specifics out of the discussion so that members can feel more comfortable without the fear that their competitors, potential clients, or website visitors will know who they are. To keep everything anonymous and encourage open, unbiased and incentive-free discussion, please do not refer to your site by name nor link to it anywhere within the Internet Marketing category. For feedback and constructive criticism on your website, please visit our Website Reviews forum within the Business Exchange category. To do business with other webmasters in our community, please visit our Webmaster Marketplace, also within the Business Exchange category.">
I honestly cant believe you cant comprehend what’s up!!
jesus christ, you are apparently an encompassed programmer related to ?
I’m flabbergasted!, yes indeed i am.
Hey sweet pee, honey of the hive, drop the spam crap, its excessive.
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/8/13 11:50 AM|
I have used the website and found it very useful... but not recently - so just re-visited....
why is the header containing an advert that has NOTHING to do with the content of the website... ie selling tickets for 'historic palaces'? and also advertising the RAC (UK car breakdown services....), health insurance.... is this a geo-targeting for adverts?
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 12:04 PM|
No, nothing like that.
She's desperate to make bucks, that's been her downfall, not Google.
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/8/13 12:09 PM|
i always use stackoverflow... never came across this website before... there is a huge difference between the two.. stackoverflow is much more helpfull... so i stick with that one.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/8/13 12:38 PM|
I see I'm not the only one drinking this fine fine afternoon.
And yeah, I love SO. I used to hit daniweb back in the day, but the popup and the poor engrish is probably why I moved on. Sometimes there's good answers there, but I fear SO has taken the idea of code help to the next level and I like it.
But I remember back in the day when....errrr dammit I forget their name! It was an IT help site that tried to change the way help site worked by a rating system. Then, they went full on paid membership and hid the answers, then they made the answer alllll the way at the bottom of the site, but they did so much deception that they never rank anymore and now they are considered a joke. DAMMIT Someone help me out here cuz it's gonna drive me nuts. I used to contribute to the site too until they got annoying about signups and subscriptions.
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 12:53 PM|
Actually in South Africa its close to 10 at night.
Drinking time!, to say the least.
So what we drinking?
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||9/8/13 12:54 PM|
You might be referring to experts exchange which worked a similar way and then became a prime source of answer bait.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/8/13 2:11 PM|
That's it, experts exchange. Hated seeing that in the SERPs because the answer was always hidden.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/8/13 2:16 PM|
YES that's the site. And, you could scroll all the way to the bottom but then they just hid it and you could see the answers in the code. Not worth it obviously. They don't rank anymore. I used to contribute there too but they got greedy too.
|Re: DaniWeb hit in November '12 and haven't recovered since||terry noname||9/8/13 2:18 PM|
Where was it hidden?
Ow, Ok, it must have been in the tittle attribute.
Reason i questioned is, i didn't see it either.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/8/13 2:39 PM|
You could open the code and see the answers. But for the longest time you just had to scroll to the bottom of the page. I guess not a lot of people knew that? lol Yeah, all you had to do was keep scrolling and the full answers were there.
|Re: DaniWeb hit in November '12 and haven't recovered since||black belt||9/8/13 2:49 PM|
The entire experience was so irritating that I simply ignored their search results and cursed under my breath every time I saw them on page one, lol.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/8/13 3:06 PM|
I only opened them when I was desperate!
|Re: DaniWeb hit in November '12 and haven't recovered since||LarryDarry||9/8/13 4:56 PM|
Sorry about that Daniweb I wasn't trying to come off like a jerk or rude. I just think you should have been smarter with your money, you didn't need an office to begin with you could have cut costs and invest like someone else said who was earning big money, they said they were earning close to a million dollars with adsense and than when the updates came their money shrunk big time and they were not longer earning a livable income but they planned ahead and put that money away invested it and didn't live lavish or overspend. One thing is for sure is Google is unpredictable we don't know what the updates bring and every time there is an update someone else gets affected.
I couldn't agree more with you, Content farms like BleacherReport and SBnation hog up those google news and search results pages all the time. I've noticed that as well when I was looking for a sports schedule there was three bleacher report articles with user generated spam one of them posting an illegal link and having thin content. Than I clicked the news section and noticed they had over 15 articles all on the first page as well as SBnation. I remember looking up some news on WWE summer slam and they hogged up the entire google news and search results. also a lot of those illegal streaming sites and forum posts hog up the results.
|LarryDarry||9/8/13 5:55 PM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/10/13 2:13 PM|
> why is the header containing an advert that has NOTHING to do with the content of the website
Must be a remnant ad. Most likely Google AdSense.
I appreciate that. I was just trying to point out that you can't judge the people behind a site based on its size alone.
The problem is that we have a $90,000 a year web hosting bill that is unavoidable. The issue with downsizing our servers when we lost our traffic is that we lost and regained our traffic three times already, and we have two-year-long contracts with our servers. In addition, they are *heavily* discounted as we were grandfathered in at old pricing many, many years ago. If we should downsize and then have to re-add the servers, we'd be paying nearly twice as much for the exact same thing as we have now. I've shopped around at a billion hosting companies, and I still currently am in negotiations with another hosting company. Unfortunately, this is an unavoidable cost.
The other problem is that we're up against sites like StackOverflow that have $20 million dollars in venture capital. And then there were the DevShed properties that were acquired by Ziff Davis Media. And all of the other competitors we've had throughout the years that had the luxury of outside investors. We never made anywhere close to any of that. It was always just me and my own wallet. However, that means that, over the course of the past eleven years, we've always needed to reinvest a lot of our gross profits just to stay even remotely competitive.
When it comes to why I had the office in the first place, it was probably a mental health thing more than anything. When you live in an apartment by yourself, it becomes almost a necessity to mental health to have a place to escape to and see other people for a bit. I live in the suburbs, so there aren't really any coworking spaces near where I live, especially 6/7 years ago when I first got the office. At the time, I was also bringing on an editor and some freelance staff writers to do product reviews along with a salesperson, so it made the most sense to have an office space where we could all work together. We were also having a lot of bulky products being shipped to us to review (computer towers, business-scale printers, etc.) and it made the most sense to have the staff writers come into the office to review the products.
We were never close to earning anywhere near a million dollars with AdSense. We just weren't. If I was, I would have put it away, and been able to live off of that after Panda hit, but we weren't making that in the first place. Not only that, but for the year following the first Panda hit, when we suddenly lost all of our traffic, just about all of our advertisers canceled on us because we were in the spotlight and getting negative publicity about our traffic situation. Therefore, we weren't anywhere close to meeting expenses. Plus, getting out of the lease wasn't exactly an easy feat either. We didn't just stop making money, but were losing money. I drained a good chunk of my savings just trying to pay hosting bills month after month.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/10/13 2:58 PM|
>> When it comes to why I had the office in the first place, it was probably a mental health thing more than anything. When you live in an apartment by yourself, it becomes almost a necessity to mental health to have a place to escape to and see other people for a bit
Can't argue with you there. It's not good living alone in an apt and never interacting with people. It's unhealthy mentally. Did it for 3 years, and my only outlet was college. If it wasn't for college, I dunno what I would have done.
|Re: DaniWeb hit in November '12 and haven't recovered since||Luke :)||9/10/13 5:23 PM|
$90k a year?
Microsoft Azure only costs 600 month and its designed for cloud companies to serve on
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/10/13 5:50 PM|
I post on a forum that claims $10k/year, but they host in Switzerland to avoid authorities. Who knows.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/10/13 6:35 PM|
We have investigated lots of cloud solutions thinking that the pay-for-what-you-use model would scale better for us since we have been gaining and losing 50% of our traffic from one day to the next. Unfortunately, after pricing out various cloud solutions compared to our self-managed servers with Softlayer (Previously The Planet), it would have cost about 50% more to switch to being in the cloud.
Luke, I suspect that is for a starter website that doesn't get our traffic levels or require our CPU power. We started out with The Planet about 8 years ago and have been with them ever since in their Dallas datacenter (they were ultimately bought by a parent company and merged into Softlayer, and Softlayer was recently bought out by IBM). Regardless, I'm lucky in that most of my pricing is grandfathered in from many, many years ago, which is why our solution is cheaper than a comparable cloud pay-for-what-you-use package would be today.
We started out paying about $8/mo back in 2002. Since then we've upgraded over eleven years to about $60/mo, then $400/mo, then $1500/mo, then $3000/mo, then $5500/mo then $7500/mo, as our traffic needs grew over the course of years.
To accommodate our traffic needs we currently have:
9 web servers (we realistically only need about 5 with our current traffic levels, but web servers are the cheapest device we pay for, and if we cancel them and ultimately re-add them, it will cost us like double, so we're just holding onto the ones we have and using them as development environments, etc)
2 database servers - we only need 1 for our traffic levels, but we want to make sure there is redundancy and a live backup failsafe in case of a hardware failure or something, to minimize downtime, since of course the database is the most precious piece of the puzzle
1 file and management server - we would need this regardless of our pre or post-google traffic issues
1 iSCSI SAN - we would need this regardless of our pre or post-google traffic issues
1 load balancer - we would need this regardless of our pre or post-google traffic issues
2 backend gigabit switches - we would need this regardless of our pre or post-google traffic issues
If we were to cut back our web servers to scale more appropriately for our current decreased traffic levels, we'd be cutting out about $10K a year. So we'd go from $90K/yr to $80K/yr ..... but then it would ultimately cost us over $100K/yr if we were to re-add them at their current pricing since we'd lose the pricing we were grandfathered in at upon canceling them.
There's really no standard answer for how much hosting costs should be. It's different for each website based on the language it's written in, the size of the database, how many concurrent visitors there are, and how active the website is (i.e. is it very read-heavy with few writes or is the database constantly being written to).
Of course, it's also about how efficient the entire system is, how much caching is done to minimize reads to the database, latency, etc. We definitely utilize memcached A LOT to eliminiate most database reads.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/10/13 6:42 PM|
> Microsoft Azure only costs 600 month and its designed for cloud companies to serve on
Firstly, we are a LAMP solution (Linux Apache MySQL PHP) so Azure won't cut it :) But let's say you take $600/mo for the average cost of a virtual machine running MS Azure in the cloud. We have 12 physical computers running over twenty virtual machines. Let's say that each of our 12 computers is in the $500-600/mo price range, but then there are additional monthly costs for our load balancer, iSCSI SAN storage device, etc. That's how it works out to about $7500/mo.
|Re: DaniWeb hit in November '12 and haven't recovered since||LarryDarry||9/10/13 7:37 PM|
Dani I feel you on the working alone and living alone, it could make anyone go mad.
Wow at 90k server fee! Thats insane. You were getting a lot of traffic but as you claim you were not raking in the millions, so you were basically taking a huge financial hit. Did you ever seek out investors or you chose not too? Ever thought of selling your site?
Most websites that get your type of traffic get paid huge through Adsense depending on the niche, but most of those sites are news sites. I don't remember the name of the site but a year or two ago I read up on a site that had similar traffic as you, but was making millions off of Adsense but wasn't paying anything close to 90k for server fees.
It kinda sucks that those YouTube make up gurus probably make more money than you posting videos about absolutely nothing useful, while you have an informative forum, but the way adsense works is on a click through basis so you can't really determine who clicks your ads. Maybe your loyal visitors are already ad blind?
Anyway hope you are able to rebound, but its pretty tough I never heard any stories of people actually recovering from panda or penguin hits.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/10/13 8:51 PM|
> Dani I feel you on the working alone and living alone, it could make anyone go mad.
That's what I'm doing now, after giving up my office. That's why I was facetiously / jokingly saying I gained 5 pounds already! Because I just sit here in my apartment day in and day out!!
I started the site while in college as a hobby in 2002, and went into it full time in 2006 as soon as I graduated. Soon after, I was being courted by a lot of venture capital firms who wanted to invest millions into it. However, I was in my very early 20s at the time, and had been dreaming the past few years about what I would do with the site once I had school done with and the time to actually invest in it. I felt like there was sooooo much I still could do with it myself, and I had my own vision, and with a computer science degree, I had the know how to make it happen. Plus, I didn't really want to work for someone else following *their* vision of where to go with *my* baby. I also am simply not a fan of the VC model. I feel like they pour millions into something that gives quick growth but is ultimately unsustainable in the long term, position it for a buyout, and then make a killing, leaving what's left of the company in ruins. I chose to turn down the VC and bootstrap it instead, and a year later, in 2008, the economy crashed, and the rest is history.
Back when we were being courted by all the venture capitalists, we had something really amazing. We were one of the first social platforms of its kind, long before Facebook or social media even existed. Nowadays, we're considered "old hat" ... forums are a dime a dozen today. Every investor I've spoken to since has basically said that they don't want to invest in an 11 year old company because if it hasn't become the next Facebook already, it's never going to.
On one hand, I could say that I passed on my one chance and it was the wrong decision. I could have been StackOverflow ten years ago. On the other hand, I could say that StackOverflow has $20 million in funding, is one of the most popular websites in the world, has strategically aligned themselves with Google, Facebook, and all the rest .......... but they have no sustainable business model and are not profitable.
The first nine years for me were amazing and I loved the ride. I got to spend every day doing what I loved and working on what started as my hobby and doing it my own way, seeing my vision come to life ... and it never stopped being my hobby. The last two years have been stressful and have burned me out because every month is a struggle to get together enough money to pay all my DaniWeb bills, my mortgage and cover my living expenses.
I mostly sell directly and have relied on AdSense less and less over the years. However, even still, forums are a tough pitch. Most community members (especially on a tech site) have banner blindness, so ad performance (and therefore ROI for the advertiser) isn't near what it is for niches with a less tech savvy audience, and that translates directly into the market bearing lower CPM rates.
Over the past two years ... We were hit by Panda, and then made a more than complete recovery. Then, we switched off of the vBulletin platform onto our own platform that I wrote myself, but all of our URLs changed, and we took a massive traffic hit from that. A second time we built traffic back up to never-before-reached highs, and then took another hit in November 2012 (not on a day that coincided with Panda/Penguin, but rather on a day that coincided with a lot of other forums and UGC sites all being hit at once, so not sure what algorithm change it was), and we haven't recovered since.
|black belt||9/10/13 11:46 PM||<This message has been deleted.>|
|mcgids n#l||9/11/13 12:00 AM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 12:15 AM|
c++ is the answer in cutting your costs quite dramatically.... at least for the webserver part... maybe you can look into hiphop to convert your code into c++
i was posting to fast based on the last comment in here.
you should always try to minimize costs and make it as efficient as possible. I understand that c++ is less easy to maintain... but as far as efficiency goes... c++ is the answer...
also look into mysql ( assuming you have mysql ) udf functions to extend mysql.
you must be hitting 20k hits a second with that setup
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||9/11/13 12:16 AM|
Dani - we host about 1/2 your traffic. (at our strong point). We use a single server, have a latency of 250ms maximum for page generation - our pages are extremely complex, even the html is parsed via a custom 'semantic' decision matrix. Our total hosting bill is 3500 GBP a year.
I've been a programmer for 30+ years, so I can talk about this with some 'authority'. You're definitely spending too much. We use different technology ofc, it's all custom built - we've now migrated the entire system to mongodb/high speed compiled IL stuff like C# (launching next year).
Anyway, hope you recover, the world deserves options. I'm on S.E. a lot (because it is obviously favoured) - not sure i've ever seen any other programming site (in the last year or so). It's a nice format, but CERTAINLY doesn't have all the answers. I'd like options, there don't seem to be any these days.
Cheers, good luck
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||9/11/13 12:23 AM|
edit: I should point out:
-We handle about 10 HTML/PAGE requests per second.
-I meant MAX 250ms for 'client page load', normally max 100ms for page generation.
-We also use CDNs like cloudflare for all image hosting (which is free), which knocks out the other 90 requests per second we would be getting.
-All our dbs are backed up, and BTSYNCd to local workstations continually.
-We used to have a backup server, but lost that due to the previous hard times that we all experienced in nov '12.
It was necessary to cut corners due to the google downfall, nobody liked it - but it had to be done.
Now we're back, we're faster, stronger and running more efficiently. What doesn't kill you etc..
|black belt||9/11/13 12:30 AM||<This message has been deleted.>|
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 12:35 AM|
i think you should be able to at least cut your costs by 50% through optimization in c++.
you can hire a c++ programmer for that or do it yourself ( you have a bachelors degree, so i am guessing you could do it yourself )...
if you hire: the roi would be 1.5-2 yr, based on the price of what a good ( excellent ) c++ coder on average costs over here.
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 12:42 AM|
ps i base the 50% on a quite similar setup and also a lamp project i converted to c++
in the above i am guessing one year job... but with hiphop... it will be months... depends on how the original php code is written ( how large ) and how much manual labor ( other optimizations, like threading ) is needed after conversion.
|Re: DaniWeb hit in November '12 and haven't recovered since||fedesasso||9/11/13 2:01 AM|
You might have missed it, but I addressed a few reasons here https://plus.google.com/114867727838406985245/posts/EBJAQruyG48
based on the examples you gave on your blog post.
TLDR: stackoverflow did a much better job with page titles.
Hope this helps
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 6:06 AM|
always funny how you get downvoted for stuff which is just true, tested and is a solution.
facebook use c++, google use c++, i use c++, its lightning fast... less cpu = less cost = happier times = higher profits = more interesting to investors.
guess c++ is not your ballgame.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 7:44 AM|
Thank you for the words of encouragement, everyone.
Over the years, I've investigated going off of a LAMP environment. Firstly, let me say that up until last year, we were based off of the vBulletin platform, so these types of extreme changes weren't plausible. I was just like every other vBulletin forum out there. Then, when I decided after our first Panda hit and recovery to write and switch to an in-house platform, I thoroughly investigated my options.
I went to a MySQL conference where I actually had a long talk (chatted with him for over an hour) with Facebook's DBA. (This was 2011). I was telling him how I was about to write a system from scratch and I wanted to investigate NoSQL solutions such as MongoDB or Cassandra instead of MySQL. He told me that *Facebook* only even uses Cassandra very little and is mostly MySQL based. I asked him if FB was slowly converting over, and he said no, that Facebook didn't feel like NoSQL solutions were mature or robust enough to replace MySQL in most cases, and so FB was choosing to only using it sparingly in places that it would give the biggest performance boost over MySQL. I went into the conference pretty assured that I wanted to learn a NoSQL database such as MongoDB and have it replace MySQL in my new system, but I left with the idea that if MySQL is good enough for Facebook, then it's good enough for me. Another factor in my decision is that I do everything myself (I run the business-side of things, I sell advertising, I do all the SEO, I manage the community, and I was the only developer.) And I had to weigh the pros and cons of taking the time to learn something new and write an entire platform from scratch based around a technology I'm not already intimately familiar with. Sometimes it just makes sense to stick to what you know if you want to get up and running quickly and need a fast turnaround. Again, that was 2011, so things could be different now for Zuck.
I also investigated Hiphop but, sad to say, most of it was over my head. I haven't looked at a single line of C++ code since college (10 years ago!) and, from what I understand, it's only meant to optimize portions of your PHP code. Additionally, we use a lot of PHP functions that I found were incompatible with Hiphop. For example, we use the CodeIgniter PHP framework and I haven't found any use cases on the web of others who are using CodeIgniter with hiphop.
We also heavily, heavily, heavily use Memcached to cache as much as we can. That's how we can survive with only one database, and we don't even use up all the resources on it fully!! (Back when we were on vBulletin, we *required* two databases, but that's because vBulletin 3.x was horribly written.)
Here's what it comes down to, though. If you go to any page on DaniWeb and scroll to the bottom, it tells you how much time it took to generate the page. On average, article pages (which have the ability to be cached) take between 0.02 seconds and 0.08 seconds, which is very speedy IMHO. Some more intensive pages like the homepage take upwards of 0.1 second, but no page on the site ever takes more than 0.3 seconds to generate.
A lot of the server-side optimizations that people consider are meant to optimize inefficient pages that take upwards of 1 second to generate. For, there really isn't much room for improvement switching to an entirely new technology. Like I said, the average cached article page (which accounts for 90% of our pageviews) takes on average 0.04 seconds to process.
It really is just that we get a *LOT* of traffic. We have lots of cheap web servers, because for redundancy and failsafe reasons, it makes more sense to have lots of really cheap web servers (that can be quickly swapped out in case of a hardware failure) than just one or two really expensive ones. And we have two databases servers also for redundancy, even though we only use one in production. Then everything else we have is infrastructure-related (load balancer, backend switches, etc.)
Also, it's comparing apples to oranges to compare how much traffic you get and how much your hosting bill is. For example, a news site won't require anywhere *near* as much resources as we do because it is mostly read-only and doesn't have a lot of database writes involved. Even more so, a website with mostly static content could EASILY live off of just one server.
|Re: DaniWeb hit in November '12 and haven't recovered since||Lysis||9/11/13 8:01 AM|
I'm working with MongoDB for the first time. It's cool because it's so forgiving, and for development it's awesome. You can develop so much easier in MongoDB, but this company takes Mongo and converts it to SQL Server (we are a Microsoft shop) for production.
I think when you're working on your bread and butter it's better to stick with what you know, and keep the experiments with small projects.
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 8:07 AM|
ok fair enough.... but 90k is still a lot of $$$'s...
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 8:17 AM|
Good news, everyone. I've recovered!!
Firstly, I want to say that the reason why I start and participate in these threads is not to complain and bitch about my traffic. I started my initial thread just like everyone else did right after the first Panda, confused about what had happened, when no one knew why they suddenly lost traffic and Google hadn't announced it yet.
I received a lot of emails and words of encouragement from people who enjoyed following my story, and said it was the only fully-documented Panda recovery story they ever read. So now I document everything I notice and everything I do in hopes that we'll fully recover yet again, and can share what we've tried and what we've learned with others out there.
For whomever it was who suggested that I remove the author metadata from post usernames, because I misunderstood Matt Cutts saying it was 'the future' instead of what to do now ... Well, it worked, almost instantaneously. We IMMEDIATELY did a 180 from a slow decline (losing about 5% of our traffic week over week for the past eight months) to a sudden 20% increase over the past two days since removing the metadata. Our US-based traffic actually increased by 100%!
We still have a *LONG LONG LONG* way to go to get back to where we started, but not only did the bleeding stop, but there was a definite 180 degree change from declining to a very clear improvement.
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 8:19 AM|
... there was an update on 4/5 september... maybe a googler over here helped as well :)
but i am glad for you. thumbs up.
|Re: DaniWeb hit in November '12 and haven't recovered since||Bigwebmaster||9/11/13 8:27 AM|
That is awesome Dani, sounds like both you and Hitchiker have recovered (or at least partially recovered), that is providing me some hope.
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/11/13 8:44 AM|
That was me. :-)
I noticed an upswing on quantcast.com for your site, actually. But it's really difficult to say if the author meta tag is the reason though. I don't know, does Google move that quickly? Who knows? Last time I checked, I was still seeing your site filtered as a blog (with nothing showing up under discussion).
It could be an entirely different reason. Regardless, glad to see some improvement. Gives hope to all. :-)
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 8:50 AM|
We were hit by something back in November 2012 that destroyed like half our traffic, and since then it's been a slow and steady decline to where we are now, at one-third the traffic we were at one year ago today.
For us, the first sign of hope in 8+ months came on Saturday, August 31st. Since that date, we have either plateaued or ever so very, very slightly increased compared to the same day the previous week. The past few days (since this past weekend) have showed even more of an improvement. For example:
Tues Oct 9th = 271K <= starting point
Tues Nov 20th = 180K
Tues Dec 18th = 137K
Tues Mar 12th = 132K
Tues July 30th = 108K
Tues Aug 6th = 106K
Tues Aug 13th = 98K
Tues Aug 20th = 97K
Tues Aug 27th = 97K
Tues Sept 3rd = 102K <= First sign of improvement in a LONG time
Tues Sept 10th = 106K <= Improvement continues
The attached image that shows US-based traffic on Mon, Sept 9th compared to the week prior is even more reason for hope.
Of course, it could be Google algorithm updates. It can also be that we reskinned the site mid-August. It can also be that we just removed the author metadata last weekend.
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/11/13 9:11 AM|
"remove the author metadata from post usernames, because I misunderstood Matt Cutts saying it was 'the future' instead of what to do now ... Well, it worked, almost instantaneously."Could this be a coincidence? Wouldn't it take at least a few days minimum for your web pages to be re-crawled etc? and unlikely to happen 'instantaneously'?
|Re: DaniWeb hit in November '12 and haven't recovered since||mcgids n#l||9/11/13 9:12 AM|
or could be that your competitors lost much of the value of links from blogs/forums/articledirectories etc. ( the ones that would normally get "disavowed" )
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 10:00 AM|
Oh, I'm sure that whatever "recovery" I've had this week is a mix of everything, most likely algorithm changes on Google's end coupled with the new skin we rolled out mid-August.
I'd like to say it's attributed to removing the rel="author" metadata. Unfortunately, even new threads posted today/yesterday are showing up in Google as blogs instead of discussions. I'm really frustrated over that!!! What can I do to convince Google that we're NOT A BLOG?!
Not only that, but now Google Webmaster Tools is showing a handful of "News" crawling errors. It's picking a sample of member profiles and giving the News Error "No sentences found" or "Article disproportionately short".
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/11/13 10:05 AM|
"Unfortunately, even new threads posted today/yesterday are showing up in Google as blogs instead of discussions."What about in the 'normal' serps?
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 10:08 AM|
Well, yes, they show up there too, of course.
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/11/13 11:32 AM|
Is it simply a 'categorising' thing?
I very rarely use any of the other search buttons apart from the 'web' one...
so does it 'matter' if a website is in web + blog or web + discussion or web + shopping or ONLY in web?
I wonder what the statistics are for usage of the various search options....
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 12:15 PM|
It was pointed out a few pages back that Google thinking we are a blog instead of discussion forums might be the reason for sending us less traffic. In other words, we have very thin content for a blog, but good content as far as forums go.
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/11/13 12:45 PM|
"Google thinking we are a blog"Just not sure how that works.... what about websites that have 'normal' web content....and a blog and/or a forum as well...
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/11/13 1:47 PM|
Or maybe it's because you have a custom CMS? Were you ever showing up under the Discussions filter?
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/11/13 1:53 PM|
We're all just theorizing, of course...
Dani pointed out that her site appeared in the Blog filter and not Discussions filter. The assumption being that Google views the site as a blog. Now, does that have an affect on overall search ranking? Who knows? If Google thinks a query is "discussion related" instead of blog related, would it push sites deemed "discussions" up to the top of the rankings?
When I run the Blog filter on my own site, it shows about 602 results; my WP dashboard says I have 699 blog posts. I have tens of thousands more pages outside of WP that make up the custom part of my site (old school).
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/11/13 8:20 PM|
> Or maybe it's because you have a custom CMS? Were you ever showing up under the Discussions filter?
It definitely used to show up under the discussions filter when we were on vBulletin. I don't recall if it did post-vBulletin. Maybe? Honestly, I don't recall for sure, but I'm pretty sure it did only because I remember optimizing the hell outta everything once we switched to the new platform, and I'm pretty sure I would have remembered if it was an issue at the time.
My frustration is that Google just doesn't understand our site at all. It thinks member profiles are blog posts, for example, which can't necessarily be good for us on the 'thin content' front.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||9/11/13 11:37 PM|
"Google thinking we are a blog"Does your site use any RSS or Atom feeds with a related update service?
|Re: DaniWeb hit in November '12 and haven't recovered since||Matti nescio||9/12/13 2:50 AM|
On Thursday, 12 September 2013 09:37:26 UTC+3, Free2Write wrote:
Being a forum , it has plenty of feeds (at least one for every sub-forum).
Some more valid than others.
http://www.daniweb.com/rss/pull/17 , http://www.daniweb.com/rss/pull/58 and http://www.daniweb.com/rss/pull/114 don't.
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/12/13 3:52 AM|
I would like to see your website....
One 'problem' with websites and technology is that over time (many years) websites can develop according to the technology that is available and of course the search engines 'may' look at the different content in different ways over that time. I have web pages within Wordpress as part of my website and Google categorizesthat part as blog content - but other websites with exactly the same niche content (it isn't identical content as such) and that content doesn't get placed in the blog section. Does this affect the long-term rankings of search terms and those pages.... in truth I don't know... BUT... my Wordpress content often gets crawled by Google within one minute and this week I was on page one of Google within that time... okay it may not be the most searched after term and it may not stay there... in fact Monday this week second on page one.. and now number 12 out of 911,000 search results (with 10 of the results above that news websites. Relevance to this discussion.... does it matter if categorised as a blog?
My non-blog content uploaded via ftp web page... can take several hours/days to be crawled unless I 'submit' a new page and then it is quicker. But there is no question that the 'blog' Wordpress content gets indexed quicker. At times I post via Wordpress - log out and go to Google to search and there is the post.... I guess Google likes the content.... it does NOT get to Bing or Yahoo that fast. Probably the two most pertinent factors are 1) Wordpress pings to Google etc that there is new content and 2) there is an RSS feed for the content - and neither of these is on the 'conventional' web pages.
The bottom line as mentioned before in this thread.. does it matter if categorised as a blog, discussion etc...
|Re: DaniWeb hit in November '12 and haven't recovered since||Suzanneh||9/12/13 4:28 AM|
>>But there is no question that the 'blog' Wordpress content gets indexed quicker. At times I post via Wordpress - log out and go to Google to search and there is the post....
Same thing happens with my posts but I'd have to disagree with the first part (that WP content gets indexed quicker). You can do a search of the forum and see lots of people who use WP and whose posts are only indexed days later. They'll even use Fetch as Google -- but that doesn't make Google index pages faster.
And indexing faster doesn't necessarily mean better ranking. Being "indexed" is just a state: "The page is in the index." Now throw in the 200+ ranking factors to rank the page.
Getting back to Dani's site, for all we know, the Blog/Discussion issue only affects the site when that filter is used. It might have nothing to do with how the pages are ranked in regular search. It *is* interesting in the sense of Google picking up signals.
|Re: DaniWeb hit in November '12 and haven't recovered since||Geminineil||9/12/13 4:36 AM|
Sorry my bad... "But there is no question that the 'blog' Wordpress content"
should be But there is no question that MY 'blog' Wordpress content gets indexed quicker and obviously doesn't mean better ranking...
I have looked elsewhere about this debate re websites seen or not seen as blogs and it has been discussed at least as far back as 2009.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||9/12/13 6:13 AM|
On the narrow issue of how a site gets classified as a blog, or more narrowly, indexed as a blog under Blog Search, the primary determining factors are the use of a feed and the use of an update or related ping service.
I doubt being classified as a blog has significant (negative) impact on organic results. But if that is the belief then, any feeds or feed services would probably need to be carefully (re)considered and administered with respect to the desired end-results.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/12/13 6:52 AM|
Thank you so much for catching the RSS bug. It's fixed now :)
Also, we use PubSubHubBub ... but so does StackOverflow.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/12/13 6:56 AM|
Oh, also, I just want to point out that in the days when we were classified as a Discussion, every listing in the SERPS said something like "3 authors, 8 posts" or something like that. Now, that meta information no longer exists in the serps since we're no longer classified as a discussion with a concept of "posts". I'm sure that went a long way to the end-user experience, and perhaps affects my bounce rate, etc.
I'm also not saying that Google is going to rank blogs above discussions, etc. But by knowing the TYPE of content it's looking at, it can judge whether the content is poor/ok/good for its kind. As someone said earlier (I forget who), good forum content is very thin content for a blog.
WORST CASE, though, even if Google doesn't take this into consideration directly, not showing the meta information about how many authors and posts within the SERPS hurts has the direct potential of hurting our searcher experience (since they don't know what they're clicking into).
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/12/13 7:40 AM|
I just found an old screenshot of how we used to show up in the SERPS compared to today.
In the past it said something like "10 posts - 10 authors - Last post Jan 1st, 2013" ..... Now, it just says something like "Jan 1st, 2011" (blog style). The problem is that before it showed a page's freshness by showing the last post in the thread. Now, it just shows the date the thread was started. We have a lot of discussions that go on for a long time, so people might not click from the SERPS thinking a page is old when in reality it's fresh content.
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||9/12/13 8:42 AM|
Not that anyone could force Google to show a result in a specific format for a specific search term, but exactly how do you want Google to show the results?
|Re: DaniWeb hit in November '12 and haven't recovered since||Free2Write||9/12/13 8:43 AM|
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/12/13 8:44 AM|
The way that they used to, and currently do for most other forums out there, including DaniWeb's sister site, www.programmingforums.org
|Re: DaniWeb hit in November '12 and haven't recovered since||julianhh||9/12/13 9:48 AM|
Stack overflow.com has 10 times as many links as Daniweb, do think that has something to do with it?
Also their links are stronger, one of daniwebs tops links is from a page within dmoz, that can't be worth much.
|Re: DaniWeb hit in November '12 and haven't recovered since||John deg||9/18/13 5:52 PM|
Dani thank you so much for posting all of this. Just to be clear - you removed rel="author" tags from the links to your member's profiles within the threads? I see that you still have "Person" metadata.
I too was affected greatly come Nov 2012 and am still trying to recover. I was also pulling out my hair a couple months ago trying to figure out why author data in search results stopped working. I eventually figured out that Google is excluding forums from displaying author data. Maybe there is some kind of weird relation here?
|Re: DaniWeb hit in November '12 and haven't recovered since||hitchhiker||9/18/13 11:32 PM|
I'm glad to hear you're recovering too Dani. Just jumped back in to say a couple of things:
1) Before 'the panda ghost update nov 2012' our traffic was around unique 180k visitors a day.
2) By Aug 6th it had slowly dropped to 80k a day.
3) After that date it began a rapid recover, we're now at 140k
I am extremely technical (not being a douche, there's a reason I mention this). After having been a developer for 30+ years, and working on the internet since 1994 - there was nothing I could see 'wrong' enough with my site to justify the traffic change. I came here and asked questions, there were some useful suggestions that I implemented. Nothing changed, but I didn't expect it would.
I removed some pages from the index (that had little impact on the site), and disavowed some links in the Goog tool.
I did absolutely nothing else. I still have author tags, that point from posts to user profile pages. And a plethora of other things that make sense. I never spam either my users or the search engines.
I must make this clear - when things like this happen, just wait. If you can see the data doesn't add up, just wait. It is possible that Google simply 'missed' something in a quiet update they did around that time. They have since corrected it.
I hope everybody affected recovers, but most of all - never depend on Google for our livelihood - it just isn't wise. I've learnt that this time around, I'll never be dependent on a single source again.
|Re: DaniWeb hit in November '12 and haven't recovered since||John deg||9/19/13 4:59 AM|
Frank, that is good advice. The strange thing with this update is it is taking a lot longer to recover than any other update anyone has ever seen. The most I've ever waited before this was 3 months. At this point I'm confident that there is nothing I can do, but I'll continue trying. Unfortunately it's hard not to rely on Google for traffic.
|Re: DaniWeb hit in November '12 and haven't recovered since||treoguy||9/19/13 12:34 PM|
Frank, that's great to hear. I've read your thread and happy to hear of your recovery.
I've posted my story here:
Did you see a gradual decline or was it a bit hit in November? Did you see any noticeable drops that would coincide with either Panda or Penguin?
Odd that your recovery happened on or around August 6th? Softening of Panda perhaps? Maybe your disavow list was processed?
Sorry for all of the questions. Trying to better understand the reasons for traffic drops and recoveries.
|Re: DaniWeb hit in November '12 and haven't recovered since||Dani of DaniWeb||9/20/13 6:01 AM|
Status Update (it's not all good!):
I removed rel="author" tags from links to member profiles within each post of a forum thread, and *replaced* them with Person metadata within each post, which I previously wasn't using. Google Webmaster Tools began to report the Person structured data type. What is interesting to note is that Google *still* classifies us as a blog. However, every page within the SERPS now says "by username" in addition to the timestamp of the first post and the appropriate breadcrumb navigation. What is unfortunate, however, is that it seems to just be using the username for the first post of the thread, and still refuses to admit that there are multiple posters per page. Google also still refuses to classify us as Discussions.
Another positive that happened (whether it was a direct result of my bitching here or not) was that I noticed that I'm starting to see more diversity in the search results. For example, the same result that was showing me Stack Overflow for 8/10 of the results is now only showing me SO for 4/10 of the results. Another benefit is that I've noticed that, according to Quantcast, some other programming forums started growing their rank at the exact same time as me. I'm very happy and excited to see more diversity and competition within my niche, compared to it *just* being Stack Overflow and *NO ONE* else being given a chance.
Then, ... the lucky streak ended. On Wednesday, September 18th, we experienced server errors along with a faulty-configured load balancer that resulted in the site being down for half a day. Unfortunately, when the site came back up, one of our six web servers was returning 404s for all requests. Not timing out, not returning nothing, but basically saying to Googlebot, "The page doesn't exist here anymore. De-index it!!"
Unfortunately, our traffic started on a downward spiral since Wednesday, and we definitely took like ten steps back. However, it seems that the traffic decrease is just for DaniWeb and not any other sites in our niche, which makes me hopeful. Honestly, I'm still excited to see that my competition is increasing rank because it really is the first time in like a year that *any* other site in the programming forums niche has increased in traffic. StackOverflow was just bringing on a slow death to the entire niche and I was desperately afraid it would be the end of the entire niche.
|Re: DaniWeb hit in November '12 and haven't recovered since||John deg||9/20/13 10:57 AM|
Thanks for the update Dani. Comparing sites in your niche is definitely a good idea.
I'm sure the traffic will pick back up once google figures out that the pages are still there.
Very strange that Google is still classifying you as a blog. Now that I think about it they removed the forum information from my listings at one point too. I have yet to figure out why. I don't believe it was in the blog style though.