Google.com SERP Changes - August 2008

[daveshap]

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

One thing I'm noticing is when I intentionally do a query that is one letter off and then click on the Google "did you mean" link, it returns one set of results, then when I click on the search button again the results are different even though the query is technically the same. The results after clicking on the search button are more in line with what I'd traditionally seen.

Perhaps Google is using different datacenters for query refinement results?

[edited by: tedster at 2:21 pm (utc) on Aug. 1, 2008]

[JoeSinkwitz]

I can dispute the reciprocal links argument; I see a few that do not engage in reciprocal links.

Edit: what I mean by this is several sites we're tracking that did fall via rerank filters are not engaged in reciprocal linking.

[edited by: JoeSinkwitz at 8:21 pm (utc) on Aug. 25, 2008]

[potentialgeek]

Robert,

Here's a video by Matt Cutts answering a question about common ownership, same IPs, same servers, and its effect on rankings:

[video.google.com...]

Summary: 2, 3, 4, 5--not a problem; 2,000--could be a problem.

But note that he says if the themes are different you need not worry. So he sort of left the door open to similar content on the same servers could get you in trouble (perhaps if the sites have other issues).

[potentialgeek]

Matt Cutts was saying to look at logs, and I'm noticing something. Google is tracking original queries followed by subsequent queries and merging the two into the final search string. No doubt it's to help Google guess what people really want to find for its suggested searches.

Example:

http://www.google.com/search?hl=en&q=keyword1+keyword2+keyword3&aq=1&oq=keyword1+keyword4

q=query
oq=original query

[edited by: tedster at 5:56 pm (utc) on Aug. 28, 2008]
[edit reason] de-link the url example [/edit]

[Keano16]

My traffic is almost 50% up in last 2-3 days... Hope it will stay :)

[indias next no1]

last 2 or 3 days down, something going on ........

[con771]

I really don't know what to say. Our traffic remains down 30% since July and we're loosing our minds trying to find a solution. We've addressed some errors from our site, increased content, links. So far no changes whatsoever.

[SEOPTI]

After testing and testing I'm now convinced the June 4 penalties were aimed to hurt cookie cutter sites. It was probably refined in July, so watch out for cookie cutter content and design within the same site or across different sites.

[edited by: SEOPTI at 1:57 pm (utc) on Aug. 28, 2008]

[clearvision]

It seems strange, we have had a slow decline and are being replaced with sites that aren't as relevant. The types of sites are news sites that has one word in the title or sites that look like they were made back in the internet "stone age".

It seems the first two entries are on topic, but after that it's anything goes.

Question: How can Google determine what a user really wants to see when the user only has what is shown to them on the 1st and possibly 2nd page?

They will pick what is presented and what has the most (seemingly) targeted title. Common sense says if you are buried on the 2nd or deeper page they don't even have a chance to "vote" on your site for that keyword term. Just a thought...

[sunroof]

SEOPTI,

Can you explain what is a "cookie cutter site" and what exactly did you mean?

[Mbwto]

SEOPTI,

We see the other way. Cookie cutter sites are coming on top. Two of our website lost some rank. Both of them are original content. One is a small website with around around 150 pages of original content and other has around 370 pages of original content. Both of this websites are very useful for the people. One of the website is very very valuable and Majority of the visitors will get satisfied when they land on this website. The quick bounce rate of most of the pages where less than 30%.

I think it is not the cookie cutter website that lost the ranks, I think it is small company website that lost rank. One of our big website gained ranking. This website has more than 200,000 pages.

[SEOPTI]

With cookie cutter sites in this case I mean yellow pages style sites where a single URL just contains the address of the company with a review form and a map.

This is just not enough content. It would be interesting to know if it is possible to get released from this 'cookie cutter penalty' by adding content.

[edited by: SEOPTI at 4:15 pm (utc) on Aug. 28, 2008]

[Mbwto]

I did some more search,

One our cookie cutter client went to 2nd page couple of months back from number 4 position, but they are now on first page again on 7th position. They have 17 unique pages and 2000 cookie cutter pages.

One of our client has a mashup site. So you may say it is cookie cutter website. But they do have more than 500 pages of unique content. They lost all their rankings. The website have more than 50K pages. The website is Mash-up site and is very informative. Visitor reviews are very positive. Even after loosing all the rank they are doing good.

I think it is something more than just cookie cutter penalty

[Mbwto]

I just checked, couple of cookie cutter websites of one of our client. It is doing good. They did not get effected.

[edited by: tedster at 9:41 pm (utc) on Aug. 28, 2008]
[edit reason] remove specifics [/edit]

[Miamacs]

Anyone noticed a temporary change in their traffic lately?
( meaning sometime this... er... I mean last week. yeah whatever: at the end of August )

I'm talking about a big but short-lived bump in the stats, that almost look like as if Google was testing something major...

...but couldn't find traces of it on the surface. Meaning it didn't affect major keywords for example.

...

If it really did happen, it was something out of reach...
way far down the long tail.

Anyone seen this?

[SEOPTI]

Mbwto, the cookie cutter penalty begins at a certain level of URLs in their index, I'm testing, but I think it starts at 5k-10k URLs.

So cookie cutter sites with just 1k URLs.. may survive.

[atlrus]

I think some tweaking is going on. To the addition of the many .co.uk domains I see, I also get A LOT of the old hacked .edu domains. Hopefully it will clear up shortly.

< continued here: [webmasterworld.com...] >

[edited by: tedster at 4:45 pm (utc) on Sep. 1, 2008]