Google Updates and SERP Changes - March 2011

[Whitey]

< continued from [webmasterworld.com...] >

< related Panda Farm Update [webmasterworld.com] >

I keep dropping mentions of this , but no takeup , so i did some digging, for clues to my theory Chrome's passing back intelligence that could influence this new algo and future changes :

New Chrome extension: block sites from Google's web search results
Monday, February 14, 2011 | 12:00 PM

Today the Google web search team launched a new Chrome extension to block low-quality sites from appearing in Google’s web search results. Read more in the post below, cross-posted from the Official Google Blog. - Ed


[chrome.blogspot.com...]

Also - [webmasterworld.com...]

I think user behaviour data is being underestimated in this thread. Each website will have an depth profile building that feeds into a potential quality assessment by Google. What say you ?

[edited by: tedster at 8:15 pm (utc) on Mar 15, 2011]

[indyank]

In google webmaster tools, i am seeing one 404 error it has 100s of internal pages linking to it. the non-existent url is "http://domain.com/forums/.

I never had forums on this site but gbot is discovering this url from many internal pages almost daily! How could this happen? I even see my home page reported as linking to it.

I even used the Fetch as googlebot tool to see whether any of these pages have any hidden link to the so called forums on my site.But I didn't find any. Why is this?

How else to find why googlebot is discovering this forums link on my site?

[snickles121]

Try using Eversoft and run the program search and find for that url for the entire website. If that url is hidden in the coding somewhere it will find it.

[indyank]

what is eversoft and where can i get it?

[tedster]

Evrsoft is a free website creator and editor - it's very easy to locate through a Bing search ;) I'm pretty sure that's the reference here. There are many other utilities that will do extend search or multiple file search, too.

[crobb305]

One thing I see that I have never noticed before today is that I am not longer being given the option to correct the spelling on a query. Google is just serving me whatever it wants, with a closest match to words in my query. I am searching for something that has the word "plain" in it, but Google is returning everything with "plan" and a lot of stuff about business plans. My query doesn't even have "business" in it. Very odd and annoying. I want what I searched for. Copy --> Paste into Bingo!

[edited by: tedster at 10:44 pm (utc) on Mar 27, 2011]
[edit reason] moved from another location [/edit]

[almighty monkey]

I'm not convinced user data is a particulary huge part of the algo.

The quality of data widely varies depending on what services both a webmaster and a user has with Google, and for all but the most popular search terms there probably aren't enough searchers to provide a large enough dataset. If a term only gets 200 searchers, and from the 10,000 results 700 websites have Google analytics and 12 users have Google toolbar installed, how on earth do you draw any sort of meaningful conclusion?

I suspect information from Analytics, Google Toolbar and the like are used when formulating changes to the algo, but are not part of the algo itself. Like, they'll test to see if an algo update was successful by looking at total bounce rates across the entire network go down.

At best, the toolbar might be used for assisting Googlebot.

edit; spelling is for the weak.

[falsepositive]

Too early to see if I'll bounce back 100% but I'm gradually crawling back from this Panda pit. Definitely seeing some kind of gradual recovery though. I have bad external duplication errors with a ton of scrapers outranking me, but have toiled heavily over a month to fix many, many things. Down 35% initially, then 2 weeks after panda, down 50% then the last 2 days seeing an inching up (marked improvement from a week ago, when it was really bad). I have not done a manual reconsideration (I was waiting to clean my entire site out completely before I went for the last resort, which was to rat out my infringers). I am going to see if I can get out of this myself without having to file a recon -- I'd like to do it as a challenge. Should be interesting! But I have hope!

Has anyone else since a gradual improvement? I MAY end today with the best numbers I've seen since Panda (totally crossing fingers). If this works and I get restored, then it will be proof that scrapers outranking is not the issue as I've only done onsite stuff. I also did a little DMCAing but it was half-hearted. I have spent 90% of my time reworking my site in a huge way though.

[Jane_Doe]

I'm not convinced user data is a particulary huge part of the algo.

I agree. In at least one niche I follow, a top spammer has sites still ranking that are pretty worthless content-wise. Interestingly, he used to have lots of worthless sites ranking, but the ones with lots of ads and the most optimization abuse are gone. But content-wise the ones that remain are no better than his others, so it leads me to suspect that user data plays less of a role than some here have suggested.

Has anyone else since a gradual improvement?

Since about early March I think all of my sites have either gone up or at least stayed the same. I saw some unusually big increases yesterday, though not on the same pages that lost traffic to Panda.

[walkman]

I rank well for some pages.

Traffic on a heavily linked section died down. I barely made any money there so am I not not that upset. With that gone it's hard to compare traffic levels where it matters.

I think it's just a tiny thing that can beat the algo, my pages are perfect as they are for users,Panda disagrees. Why add 2-3 paragraphs for no reason?

Google.ca now makes 20% of my Google traffic, no Panda there I suppose.

[tedster]

I don't think every SERP change coming out of Mountain View is related to Panda - except that Panda is now a factor folded into the full algorithm, from now on. But from the discussion here, it seems like everything is being seen through Panda glasses.

For example, a site I've been working with for 14 years is now seeing long time detractor sites jump from page 2 to page 1 - even though those pages are unchanged in five years and have not received new backlinks.

I'm wondering about this in relation to a recent change where the word "scam" no longer triggers auto-suggestions. You now need to press the search button to signal Google that you really want to see those results.

This is a good change, as far as Suggestions go, but did Google over-compensate for it and boost detractor sites to somehow balance things out?

[ckissi]

@falsepositive, is this US traffic that recovers ? because I see the huge jump too but my is mostly from India, I see no improvement from US (and CA, UK ...) so far

[falsepositive]

@ckissi, A closer look and I'm seeing a big jump from Canada and other countries. US still the same... Ugh. And I was feeling hopeful too!

[outland88]

Where are those traffic throttling posters like Drall?

[crobb305]

A couple of questions I am pondering.

1) Do the class names in CSS carry any "seo" weight? For simplicity, I created a class name for a CSS rollover button. The class name happens to be a phrase I was demoted on. I am wondering if I should change that class name. To help keep my CSS organized, I named the class "widgetrollover", but I can't imagine it would be harmful. If "widget" has been over used on the page, triggering a phrase-based re-ranking, I wonder if the class name should be renamed or am I going overboard here?

2) Regarding the rollover buttons mentioned above, could the style (i.e., font size and weight) carry any seo influence? For example, if the CSS rollover button uses a 14pt bold font that jumps to 16pt on rollover, could that contribute to any spam/OOP penalties? I just added these in January to make them more visible to the user and to eliminate slow-loading images. My concern is this might be pushing my "spam" score over the top, given that the phrases in the buttons happen to be the ones that got demoted. Maybe I should revert back to small text/or images?

[zerillos]

@crobb
You're not going overboard with that. I suspect an influence too. If you get to any conclusion with that, pls share.
The reason i believe styles may have an impact is because i observed they have a negative influence on adsense targeting when inserted directly into the html of the page, instead of a separate css.

[crobb305]

Zerillos,

I will let you know how my experiment goes (reverting back to image buttons). In January, in an attempt to speed up the page load time, I removed image links. I created CSS rollover buttons (logical solution) to 4 internal pages. For the CSS class name, I thought it would be simplest just to call it "widgetrollover". Now, the other concern is that suddenly I have additional anchor variations (phrases now penalized) linking to those pages, whereas before they were images.

I have determined my penalty to be very phrase specific, so now I may have to revert back to images and eliminate the CSS rollover. This is unfortunate for the user in that it will increase page load times. But it's clear, we no longer build for the user. We are now building for phrase-based considerations.

[zerillos]

Why should phrase variation be considered evil? If all the anchor texts stick to the subject of the target page, and G can very well determine that, then there shouln't be a problem. Linking to a page about balls with an anchor about plastic is not ok, but linking to the same page with 'soccer balls' and 'golf balls' seems perfectly legitimate to me.

If you revert and it works ok, i belive it's because you have too many links to those particular pages. In my opinion, image only links don't have the same weight as text
ones. Or maybe too many links on the 'source of the links' page...

For some reason i always chose not to include keywords in the css class names. It looks to me a bit black hat and keyword stuffing. I don't really have a reason why. It just does. I always wondered if using keyword branded css class names really does have any seo value. If it does, than OOP could apply to them too.

[outland88]

How are the pages ranking in Bing Crobb or did they loose the rankings there to?

[crobb305]

Outland88, my pages in Bing/Yahoo rank well. If anything, I have seen a slight improvement in recent weeks (except on one phrase -- see next paragraph), probably due to the addition of new content post-Panda. There is no issue of scrapers outranking me for any of my content whatsoever. Google is a scraper's haven.

I did see a slight drop in a key phrase on Bing (from position 2 to position 5), but I think this is because I removed the H1 tag around that phrase (on the homepage) to sort of deoptimize for Google. I added the header tag in January (my first use of header tags on the homepage of this 7-yr old site), so I felt it may have raised a red flag for OOP.

I am just taking stabs in the dark really, essentially undoing the tweaks I made in January, rewriting content that has been scraped by lazy leeches, and adding content to the thin pages that took the hardest hit. I am studying the phrase-based scoring patent, and identifying potential problems with my internal linkage and overall wording.

So, long story short -- traffic is still good, but a slight drop on Bing for one phrase by removing H1. It is a sad sad day for webmasters. No more building sites/content for users. Fortunately, Bing doesn't seem so hung up on penalty, and they are competent at rather accurately identifying the original among scraped documents. I like their (Microsoft's) website-credibility study, and I am trying to glean some information from that as well.

[zerillos]

Here's an interesting situation.
An article i wrote a few days back does not show up in SERPs anymore when searching for title. If i search for "title" (between quotes) it's on the first position.

I searched again for title (without the quotes) and still doesn't show. I switched to 100 results per page, went to the last page and clicked "repeat the search with the omitted results included." - my article was on the first position.

any ideas why this would happen?

thanks!

[falsepositive]

Amazing. I wrote a huge article yesterday and now aggregators beat me in every phrase combination I can think of for this article. The aggregators sport only snippets (because I switched to RSS excerpts). They all link back. I've cleaned out my site and made a lot of great improvements. But Google still decides to rank the aggregators above the original pieces (in many cases, several places above the original article). Seems like with a set up like this, you might as well just give up doing anything.... I have to say that this algo is a huge FAIL.

We're all working hard, making great strides towards quality -- yet, no real solid indication that we are on track or we are being rewarded for the fixes. AND the real insult here is how the BAD GUYs continue to proliferate at our expense. A sad day for webmasters indeed.

As an aside, my business partners are already rumbling and wondering about breaking off deals and partnerships. Google has destroyed the small guy with this one unless we start seeing improvements soon.

[ckissi]

@falsepositive, everything changed about 5 hours ago. Within a hour I lost all "plus" traffic I got in last 2 weeks. Seems to be about the same as 2 weeks ago before I noticed the increase (or even less a bit).

I don't know if you have online stats for your sites because I really wonder if you or others here experienced the same.

[backdraft7]

After a welcomed leveling out of traffic and sales in the past weeks, yesterday at about noon CST everything just dropped out. Not sure what's going on...I guess I need to read backwards to discover the latest mess. Traffic is there somewhat, but again, being redirected to sub pages, not my main entry pages.
Big up & down spikes too.

[edited by: backdraft7 at 3:42 pm (utc) on Mar 30, 2011]

[falsepositive]

@ckissi, my Google US traffic is the same. I am seeing an increase in foreign traffic, which to me is pretty irrelevant (US is my business). I am seeing traffic from scrapers, which has become more noticeable. My overall traffic is higher than a week or two ago by 15% to 20%. But the killer is that I've done what I feel is a tremendous amount of work over the last 4 weeks. I've toiled to cover all the bases: technical issues, design issues, content issues all to address Panda quality signals and criteria. While I've been rewarded with some improvement, I'm still seeing the sitewide devaluation do its thing as new posts are buried under the weight of scrapers. So that aspect is still my wildcard and has been very difficult to address.

After doing all we can (except fight scrapers), I just want to ask Google: what more do you want from us? My final resort IS the manual review.

[ckissi]

@falsepositive, I doubt that this increase has nothing to do with changes on sites, because I didn't change anything and got also 20% increase. Some other webmasters here also mentioned the same increase and they didn't make any changes on sites.

As you said US traffic = no change at all. It was only international traffic for me that increased, but as I said it's gone already. Started to drop at 10:00AM GMT for me and I'm back to post Panda completely now.

[falsepositive]

FWIW, does anyone get the feeling that this algo is not really that well received in general/overall? Not sure if it's just how these changes work, but the vibe I'm getting is quite negative about this update. This is the general case even in my own niche, where even the "winners" acknowledge that there seems to be something wrong with the rerankings. Many of these webmasters admit that good sites were unjustly hammered while mediocre sites were promoted. I am also noticing more people getting more agitated over this. The more I hear about about the scraper issue, the more afraid I get -- I am just now hearing about how isolated webmasters have battled this situation before and have had a very hard time with it. Things clear up only after a long stretch of time and after much effort is made to address it. I used to be quite nonchalant about scrapers but that's changed now (color me oblivious on this matter). Such a shame, since I used to spend my time truly catering to my users and now will have to devote time to policing my own site as well.

[tedster]

does anyone get the feeling that this algo is not really that well received in general/overall?

Not really. Some site owners seem to be having problems, but I still think it's a vocal few. And the average search user? We'll see when the March market share stats are released, but I'm not hearing the same level of criticism that I did in Jan-Feb from people I know whose job requires a lot of research using Google.

Site owners never like ANY algo change that hurts their traffic, and that's been true ever since Google became the dominant search traffic provider.

[falsepositive]

Thanks Tedster. It's my first serious slap from Google so I guess I've been way too spoiled over the years. In a way, it's a reality check and I feel like I'm going through the kind of initiation that you all have already experienced as elders in this line of work. ;-)

It would probably benefit us newbies to see how things unfolded in past cycles. It's rather frightening to hear about business wipe outs that came with this kind of change but I'm also still hopeful (despite the evident frustration) that this will eventually shake out in our favor.

At the same time, I just never realized what a sordid mess the duplication issue has become. Never seen it worse than it is now -- I'm just afraid a lot more scrapers are seeing this as an even greater opportunity to feed off of stricken websites already stunned and made vulnerable by Panda.

Think about it: if I were an unscrupulous G gamer, all I'd do is look for sites that were hit, know that they can't rank, feed off their content and there may just be a chance I'd outrank them. It's like the vultures are out and about!

[tedster]

We can look at the Florida Update [webmasterworld.com] in 2003 - that was pretty earth shaking for many site owners. As I recall, the tumult was even more extensive than it is with Panda. Panda affected 12% of all searches, but Florida seemed to be a lot more than that.

With Florida after an outrageous amount of posting, complaining and confusion, two things happened. The first was that site owners understood something new, and the second was that Google continued to tweak things and the harshness seemed to get more mellow for many people. I'm sure we'll be seeing both those elements emerge with Panda. Well, maybe not the mellowing so much ;)

[outland88]

I would think the market share stats would be heavily skewed in Google’s favor. The simple reason why is webmasters are pounding search to regain rankings. It's probably equivalent to the Florida update because more people than ever do business on the Internet. The SE -Land research showed who the winners and losers were and it mainly seemed to be big business and branding. A mountain of money was shifting hands. I really can’t surmise as to what you’re hoping to see or expect to see in the future stats.