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]

[Panthro]

If GOOG is using semantics to determine if content is spam, I can see a lot of forums and other discussion websites taking a hit when topics are discussed by users whose primary language is not English.

[TheMadScientist]

That has been exactly my point in some of the other threads.

You're only looking at the page that dropped ... When you start looking at the page(s) they were replaced by you know it was a re-ranking, not a penalty. I fairly sure it's easy to understand if you read my previous post.

[browsee]

"sites with original content and information such as research, in-depth reports, thoughtful analysis and so on"

I think it makes sense why Squidoo is doing better than HubPages. G changed the algorithm to see data, tables, images, comments, videos, qna etc on the same page instead of 10 different pages without a canonical tag.

[browsee]

“We’ve become a civilian casualty in the war against content farms...Why us? We have no idea. The changes Google has made to its system are secret. What makes it worse is that Google’s tinkering seems to have actually improved Demand Media’s page rank, while killing ours...We’re a blog, so we aggregate news stories like everyone else. But our posts are 100% original and we do a ton of original reporting...” - from [cultofmac.com...]

read the words "civilian casualty in the war". I feel exactly the same.

[AlyssaS]

Here's a question for those who have gotten hit - do you have Adsense on your site?

From everything I've seen and heard, most of the sites hit have had Adsense involved somehow - either the backlinks were from sites that had adsense and were devalued, or the site itself has Adsense and has been deemed a farm.

My main site (which doesn't have Adsense on it) has moved up five places for it's main keyword. BUT - a couple of pages which I wanted to boost, and which I had written articles on Ezine and other places linking to them, have dropped back to the position where they were before I wrote the articles. In other words, I might as well have not bothered.

Same story on my other sites. Site #2 has stayed in exactly the same place for keywords where the backlinks came from non-Adsense sites, but where the keywords had anchored backlinks on Adsense-monetised article sites, they have fallen back.

If you look at the list of the sites that got penalised, they are plastered with Adsense. Hubpages and Ezinearticles put adsense within the articles, but Squidoo, which escaped, does not.

At the end of the day, in terms of data collection, Google knows more about sites that have Adsense than any other type of site, because of the tracking - and the ad placement gives them an idea of where the user is clicking out. If people are consistently clicking out on the ads within the first paragraph of an article, is this an indication that they've been turned off by the very first few sentences, and decided the article is not worth reading in full and they'd be best off finding another source of info and clicking out? Is that the mark of a junk article?

[universetoday]

There are lots of high quality sites that don't have Adsense that still dropped.

[chrisv1963]

The last couple of hours the SERPS have been changing constantly for some of the keywords I'm monitoring. Does anyone see the same?

[AlyssaS]

There are lots of high quality sites that don't have Adsense that still dropped.

But what about their backlinks? Are they from places with Adsense on? If the backlink gets devalued, down you go in a domino fashion, even if your site is all lovely and good. Or maybe the site linking to you has no Adsense, but their backlinks got devalued because they are from article sites, and a domino effect occurs.

There's a lot of dominos toppling throughout the SERPs at the moment.

[AlyssaS]

Are people familiar with the six degrees of Kevin Bacon game? What we might be seeing is a six degrees of Ezinearticles SERPs. If you are in any way related, you get sucked down.

[ckissi]

@chrisv1963 I'm experiencing something like that. Seems to be some constant resort because my hourly stats shows bigger variations as usually.

[GeraniumV]

After some more analysis - it appears that my traffic has seen a lot of chopiness the whole of February - which is peculiar as you can usually set a calendar by it (I look at the year as the carrier wave and the week as the modulated signal riding on top of it).
Drops and surges all over the place.
Looking at keywords and landing pages it appears that the primary effect is a site demotion - although I'm sure that there are more factor in play.

[chrisv1963]

because my hourly stats shows bigger variations as usually

Same here. Actually I saw such unusual variations too the day before the big update.

[pontifex]

AlyssaS: right, that link profile evaluation is probably the big mystery here... if C gets major link juice from B, but B gets power from (a1, a2, a3) and that was just a chain of "farmers" or is now devalued in the eyes of Google, C is suffering here and that is a tragic victim of basically friendly fire, if C was competing for traffic from a1..3

That expresses the complexity of that update analysis and I still do not see where to counteract precisely... Another thing that puzzles me is (especially in my field): if you have a product

"car sound effect" to download or "lady gaga - bad romance" as MP3, what expects Google from me?

Shall I mix in some youtube stuff, RSS feeds and 2 free wallpapers from lady gaga or 5 car pictures from flickr with text to countermeasure "thin content"? Then I would be a farmer, too and that is not wanted either, right?

It is just a sound effect of a car. Not much to say about that file... 2 lines of text, maybe 3 (max!) - everything on top would be for search engines only.

Further: Commercial search terms are IMHO a completely different story than information searches or even science terms.

Changing the algo to "one size fits all" and turn the whole economy around is a big stunt and I sure hope they know what they are doing. From where I started to search deeper, it sure does not look better in the long tail.

There are major scrapers and content farmers gone with the major terms - but did you look at 3-word or 4-word searches a lot? That long tail is dirty or scraped or both and still not right (IMHO)!

Funny thing is, the domains in these results I checked today LOOK very nice and I found 3-4 guys who must have a HUGE grin on their face now!

I stumbled into 2 suddenly high ranking forums - sweat layout, lots of entries - BUT every further click leads into pop-up hell of newsletter signups...

Nice try, Google - so far I am not impressed, though. The stuff up there does look better on the first glance, but maybe someone at the plex should actually try to use the results, not just look at the landing page if the layout feels right!

P!

[ckissi]

chrisv1963 , seemed to me like US traffic raised then dropped again, definitely some resort is in progress

[dickbaker]

You're only looking at the page that dropped ... When you start looking at the page(s) they were replaced by you know it was a re-ranking, not a penalty. I fairly sure it's easy to understand if you read my previous post.

TheMadScientist, I understand what you're saying about the pages that replace pages that dropped. I can look at them and say, "okay, they're relevant, and there's something Google considers good about them. Let's examine them."

What I do not understand is the degree of the drop for those pages that dropped or, if we're to look at it only in terms of re-ranking, how pages that are almost or completely unrelated to the query can be ranking somewhere between #20 to #48, and the page that had been on page one and was considered relevant and good is now placed at #49? And what of the page that was considered relevant and good for that query phrase now not even appearing at all in the results for a search on that phrase?

[falsepositive]

dickbaker, here's one example I have. If it's not a sitewide issue, I went to spot check some of my pages.

I went to check one of my top pages that was reranked -- from #1 for a solid short phrase that is not very competitive. I puzzled over why it fell to #3.

The page is clearly relevant to the keyword, and is why it was #1 for years. The new #1 position is occupied by Wikipedia, that has a few sentences for a MOVIE that has the same phrase, that looks to be completely out of context with the rest of the content that you'll see on the google search page. The #2 position is by a spanish site (it's all in spanish) using an URL which has the english phrase in it.

My page, which has a ton of comments, discussion, pictures, tips, etc. is #3.

I checked my page in copyscape and it has been scraped to the HILT. It was such a popular page that it has been circulated everywhere, from blogs, to big sites, everywhere.

Anyway, that's just one example I'm seeing. The overall result for me is that dreaded reranking loss (although it sure feels like a sitewide issue, as I mentioned earlier).

Missing pages I have seem to show a multitude of clones that scraped it. But I've also got some missing pages on certain keywords that also don't show clones. So not sure where the pattern is with that.

It's possible that on one case, it is because of dupes. The other case, it is just because they are less relevant or have weaker links.

[TheMadScientist]

was considered relevant
...
And what of the page that was considered relevant and good for that query phrase now not even appearing at all in the results for a search on that phrase?

Because the relevance score wasn't changed afaik.

Quality was added in ... The drop does not mean the page is no longer relevant, but rather it means the page does not contain the perceived quality of other results, imo.

If you're looking at relevance signals to regain rankings, imo you're looking into a black hole. Quality Scoring was thrown into the mix with everything else, so that's where I would start looking to find recovery.

Throw the keywords and link(ed) text out of this one for a minute and I think you'll do better on tracking things down. Think about 'foot prints' and 'patterns' related to the document (page(s), site(s)) rather than the formerly all important links and topicality.

This update was totally different...

[ponyboy96]

I'm thinking this was more of a re-ranking and re-scoring update. I'm still collecting data around this, but from what I'm seeing in the rankings there were some slight shifts across all sites that I manage. Couple of winners, some losers, and one big drop for one.

These shifts are slight, only a couple of spots on average, but considering some of these have been very consistent in rankings for years, it's alarming.

[edited by: ponyboy96 at 7:49 pm (utc) on Mar 1, 2011]

[TheMadScientist]

Oh, ponyboy96 I think you might be on to something...

I started the following thread and tedster helped translate my post to English for everyone. Hope it helps out with understanding why looking for a penalty probably won't provide the answer.

Google's AdSense Farm Update Was a Re-ranking - NOT a Penalty [webmasterworld.com]

[rowtc2]

In top results are 2-3 pages from a single domain (a third party authority, not the brand name) instead the previous diversity.

Instead of dealing with multiple individuals, they have choosen the simple way. Put in top ten 2-3 results from brand name, 2-3 results from an authority domain, 1 Wikipedia, 1 Youtube, 1 news and that's it. Anyone can say results are not relevant? Great job engineers.

Blogs and independent articles was welcome to be an alternative for the surfers. People may check different sites from results to see different opinions, to be informed. I think the Internet has expanded too much they can deal (including the percent of spam) and has choosen a simple method to resolve it. A lot of damage to good faith sites.

Domain authority + page authority and relevance (+ user behavior if they can implement) is enough to make a difference between normal result and spam. New rseults are not based on quality, just put it in front only big authority sites.

A lot of my pages are now on 30+ position instead top 10, what was good is now garbage. Useful landing pages, what users are looking for, a lot of value added.

Are the new results based on quality? Are diverse? I have doubts, for me is a step back in time.

[zoltan]

rowtc2, I see exactly the same pattern. However, we might be subjective. Let's wait for the mass to comment on these changes.

[vandread]

Why would Google deliberately go after Adsense sites? That makes no sense at all.

[kd454]

"In top results are 2-3 pages from a single domain (a third party authority, not the brand name) instead the previous diversity."

Seeing lots of this going on, look at health related areas, lots of complete change ups, unless your a newer site which all of my newer sites were untouched, they also have the best quality writers I have ever used, maybe a correlation.

[netmeg]

Why would Google deliberately go after Adsense sites? That makes no sense at all.

Not ALL AdSense sites. Just those of a certain type, maybe.

[ckissi]

They're definitely doing something , more US referrers in online stats. Anybody noticed the same ?

[dickbaker]

If this is indeed what The Mad Scientist and others believe it to be--a whole new way of evaluating pages for ranking, it presents a conundrum. The pages I have that were ranking well on Google are still ranking excellently on Yahoo, Bing and other engines. In fact, they're ranking the way Google used to rank them a year ago. So, making changes to satisfy Google could jeopardize the Yahoo and Bing results.

That means putting all of the proverbial eggs in the Google basket.

[zerillos]

AdSense by itself has nothing to do with the drop and i have several examples to support that. However, many people consider sites with adsense as having low quality.

As many have mentioned, google doesn't word its policy as going after 'relevance', but after quality. I might just have an idea on what's this all about. I hope I'm not right because it would mean the SERPs will just become dumber. In my niche, i'm seeing pages with lots of useful info getting replaced by two sentence blog posts...

[Whitey]

I have hundreds of FB "likes" spread over many pages, and I got hit. Competitors with no "likes" did not. So what about those signals?

I've not seen any evidence that FB "likes" currently contribute to the algo quality score. No doubt there were other overiding signals that you are juggling with.

Maybe Google hasn't flexed it's muscle on it's Chrome analysis yet, ( or maybe it has ), but the possibility has to be brewing strongly ... still waiting for those takers inputs :)

[Lorel]

I've never seen Quantcast being anywhere near actual traffic, and the same with similar sites.

[universetoday]

Quantcast is very accurate with registered sites. I can verify that. Unregistered sites are hit or miss.