homepage Welcome to WebmasterWorld Guest from 23.22.194.120
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Home / Forums Index / Google / Google SEO News and Discussion
Forum Library, Charter, Moderators: Robert Charlton & aakk9999 & brotherhood of lan & goodroi

Google SEO News and Discussion Forum

    
two word search terms
big change in Google serps
annej




msg:714940
 8:52 am on Jun 5, 2005 (gmt 0)


I've been trying different 2-3 word search terms that always used to bring up related pages on my site.

The change is unreal. For example searching the three word name of a woman in history I get several pages for results showing only one or two of the words in her name and these scattered over the page. There is only one result with her full name. When I put her full name in parenthesis then I get 4 results. A page written by another historian that mentions her name in a link, 2 scraper sites who copied a snippit from my site and last is my article. Thank goodness Yahoo still puts my article on top. (without need for parenthesis)

I just don't know how to deal with this sort of thing. Always before there was something I could do like make my title clearer, see if anyone would be interested in linking to it, etc. On this one I feel helpless. I can't imagine what to do. Any ideas? There must be a way to cope with the Bourbon serps.

 

MyWifeSays




msg:714941
 4:28 pm on Jun 5, 2005 (gmt 0)

I've noticed this too. The proximity of individual terms and the number of times each individual term appears on a page seem to have dropped in importance.

The result being that far more pages qualify for the search term, thereby increasing the competition to include top sites (e.g. BBC) which as you say may only have a single occurrence of an individual term (or terms) but score highly on other factors like PR, site rank and number of links.

I believe this to be a major factor for many pages dropping in this Bourbon update. It's not so noticeable for competitive phrases because many high ranking sites compete for the complete phrase.

John Carpenter




msg:714942
 8:55 pm on Jun 5, 2005 (gmt 0)

When I put her full name in parenthesis then I get 4 results.

You might want to try using quotation marks instead of parenthesis.

annej




msg:714943
 9:25 pm on Jun 5, 2005 (gmt 0)

What was I thinking! I meant quotation marks. I've gotta start rereading what I write before I post.

bether2




msg:714944
 11:25 pm on Jun 5, 2005 (gmt 0)

I can't imagine what to do. Any ideas? There must be a way to cope with the Bourbon serps.

I'd wait at least a week before doing anything. I believe googleguy said that the update won't be complete before then.

johnhh




msg:714945
 11:37 pm on Jun 5, 2005 (gmt 0)

We have noticed this as well on three word phrases. I have some tests running - of the pages recached most ( but oddly not all ) have improved position.

Short titles matching search phrases exactly are OK AS LONG AS the words are shown in the page 2 or 3 tmes.

Earlier this week Google was picking up key words in URL links on scraper and directory sites. This seems too have stopped now.

annej




msg:714946
 6:36 pm on Jun 6, 2005 (gmt 0)

I know people can always search using quotes but I don't think many people in the general public do. It seems to me weight should be given to web pages that contain the full phrase. In three word searches some of the top 10 results don't even have all three. It's like a giant step backwards in serps.

Hardwood Guy




msg:714947
 6:40 pm on Jun 6, 2005 (gmt 0)

"but I don't think many people in the general public do."

Count me in as one who doesn't use quotes and I don't consider myself general.

Atticus




msg:714948
 6:56 pm on Jun 6, 2005 (gmt 0)

Google's inabilty to match relevant, multi-word phrases began at least a year ago and has increased over time. It looks to me as if G is using a "too relevant" filter. I have no idea how it works and whether it is a bug or a feature.

I specifically asked about this issue in the "Ask GoogleGuy" thread, but no answer has been forthcoming.

This bug/feature makes life difficult both for searchers and publishers.

MyWifeSays




msg:714949
 8:09 pm on Jun 6, 2005 (gmt 0)

Yes, I first noticed it about a year ago after one of the updates. People then were talking about 'over optimisation penalties'.

It only lasted about 3 weeks in as severe a form as we are seeing now. I haven't noticed it since then.

Of course another interpretation would be that the algo has always been the same in this respect, but it's just other factors which have become more important, such as a rank given to a site as a whole.

Whatever it is, the results are suffering and I can't think of any reason for it other than change for change's sake.

WhoopsAccident




msg:714950
 11:15 pm on Jun 10, 2005 (gmt 0)

This is the search term that took 3 days to get to the tope of Google and then disappeared on 21st May (4th day):

cooking holidays in the dordogne

If I search for it now the result is position 101.

If I write it in quotations like this:

"cooking holidays in the dordogne"

Then I am position number 2.

I don't think mosts surfers even know about using quotation marks so that does not help. I also believe from my own experience that phrases have only been hit since 21st May.

I have examples of Googles search results not been relevant which I have referred to Google via someone I was in contact with at the company. They ignore what I put forward as hard facst and just recite the same old trite statements:

"As you may know, as we add new pages and incorporate updates to existing pages, you may see changes in the ranking and inclusion of sites in our index. Some will be ranked higher than before for a particular keyword and others lower. Because our index changes regularly, it's possible your site will regain its ranking."

Extremely patronising. Why can't they answer the question - usually people avoid the issue / the truth when something is seriously wrong.

I think some of the search results that are coming up are complete . . . let's say rubbish. Wasn't this waht killed Alta Vista as the original leading search engine?

WhoopsAccident




msg:714951
 11:18 pm on Jun 10, 2005 (gmt 0)

. . . I also meant to say that Yahoo has it right. I list at number 4 and 6.

I am fast going off Google and hope that the rest of the world does also.

{:-))

sailorjwd




msg:714952
 11:38 pm on Jun 10, 2005 (gmt 0)

I want to clarify the answer I got from Google concerning Penalty for 4-word search which is exact company name.

I hope I don't get in trouble here for quoting then exactly:

'f... off'

annej




msg:714953
 1:10 am on Jun 11, 2005 (gmt 0)

I just noticed something interesting. The homepage of the site that plunged so low comes out at #10 on the term (I'll call here) widgeting circles. The 'ing' being the difference. It used to be 98 for plain widget circles but now is lost in oblivion (used to always be 2-4). The widgeting phrase is not even in the title though it occurs on the page. Most if not all (I haven't tried them all) of the pages on the site are also lost or several pages deep in the serps. I mentioned before the thing I feel the worse about is I have several article by top people in the field and all that good, well researched information, is now lost to students and teachers. I think it's really odd that the homepage is showing up so well on one phrase when all else is lost. Any ideas why?

An amusing side note is that serp #9 on the 'ing' version is the Google directory which places this site 4th under the topic.

oldpro




msg:714954
 6:42 am on Jun 11, 2005 (gmt 0)

It looks to me as if G is using a "too relevant" filter. I have no idea how it works and whether it is a bug or a feature.

I agree. What I am seeing is sites combining two word phrases such as "blue widgets" are ranking well below sites that have "blue" and "widgets" mentioned separately within the content.

Another commonality I am seeing in the various posts here is that sites ranking well on yahoo or msn, rank poorly on google. Maybe google is just trying to be different, or either their reliance on off page factors such as PR has backfired with regard to relevancy.

Who knows? A famous political historian once said, 'unchecked democracy soon leads to anarchy'. Page rank was google's attempt to become the "democratic" search engine. The question now is how do we seo for anarchy...

"widgets blue", "blue that are sometimes widgets", "widgets that have the blues"?

Bourbon...Google's journey to the heart of darkness..."the horror, the horror"

Just a little comic relief folks.

ncgimaker




msg:714955
 8:49 am on Jun 11, 2005 (gmt 0)

WhoopsAccident, I notice with the quote search that another site comes above you, the article is exactly the same, the pictures identical.

Without quotes, that other site comes 4th. I think this is what G were after with the duplication algo.

From Googles perspective there is little point in displaying the same article multiple times in the SERPs, so your copy of the article may be being pushed down the rankings to avoid that.

If you can outrank the 'french' site, my guess is he would drop down to be below 100 instead of you.

JuniorOptimizer




msg:714956
 10:33 am on Jun 11, 2005 (gmt 0)

I've been plagued by the "page too perfect" filter since the 2/2 update.

Google has given up on relevancy, what does that tell you?

sailorjwd




msg:714957
 12:38 am on Jun 12, 2005 (gmt 0)

kiss of death to have page title and H1 tag having same keywords.

BillyS




msg:714958
 12:59 am on Jun 12, 2005 (gmt 0)

kiss of death to have page title and H1 tag having same keywords.

Two observations. I posted about a month ago that it seems a search in quotations brings up better results in Google than without the quotes. My take is that in their battle against spam they are overshooting the target.

Page title and H1 tags being the same may be considered over optimization, but it's really just good practice. This is what W3 has to say about H1:

<h1> is the HTML element for the first-level heading of a document:

If the document is basically stand-alone, for example Things to See and Do in Geneva, the top-level heading is probably the same as the title.

In fact, if you visit W3 you will see that they follow their own rules and make the Title and H1 the same. It's interesting to think that some believe that Google might penalize for good practice.

oldpro




msg:714959
 1:05 am on Jun 12, 2005 (gmt 0)

kiss of death to have page title and H1 tag having same keywords.

So...are you saying that if your title is Red Widgets From Mom & Pop Manufacturing...

Your h1 cannot read Red Widgets at Wholesale Prices?

If this is true, too many do's and don't's and what was a do a month ago is a don't today.

I kindly ask...cite your evidence.

sailorjwd




msg:714960
 1:25 am on Jun 12, 2005 (gmt 0)

oldpro.

Don't have hard evidence.

I'm looking at my niche search for microsoft access consultancy company and find that the 1st result has an on-page keyword density for microsoft access of zero.

When I search for exact titles of any of my pages (given I think my entire site as a spam penalty even though G says not) I don't see results with those words in the titles even though there are tons of them out there.

<added>
It is likely that this isn't always the case but maybe just on phrases that Google sees as Money phrase based on Adsense and scraper frequency.
</added>
It seems that for 3,4,5 word search you need a near zero keyword density to come up to the top.

This would certaining kill a lot of spam sites and mine complied with their tactics in terms of high density... page filename, title, h1, backlink to home page, first paragraph words.

For my pages that have a little meat in them I still rank often 1 & 2 on yahoo.

WhoopsAccident




msg:714961
 1:48 am on Jun 12, 2005 (gmt 0)

"WhoopsAccident, I notice with the quote search that another site comes above you, the article is exactly the same, the pictures identical.

Without quotes, that other site comes 4th. I think this is what G were after with the duplication algo.

From Googles perspective there is little point in displaying the same article multiple times in the SERPs, so your copy of the article may be being pushed down the rankings to avoid that.

If you can outrank the 'french' site, my guess is he would drop down to be below 100 instead of you."

This article was reproduced with the permission of the author. No different to the real world of publishing.

I agree with what you are saying but why did everything change on 21st May.

Furthermore if the original publisher of the article sits at number 4 on Google iwthout the quotation marks why does this sit at number 1:

[frenchconnections.co.uk...]

caveman




msg:714962
 4:20 am on Jun 13, 2005 (gmt 0)

... a month ago that it seems a search in quotations brings up better results in Google than without the quotes. My take is that in their battle against spam they are overshooting the target.

Bingo. That, plus LSI tweaks, plus redirect and canonical issues = 97% of the muck ups in this update.

But the one BillyS calls to our attention was by choice, I think. Ill conceived as it may be.

ncgimaker




msg:714963
 7:23 am on Jun 13, 2005 (gmt 0)

Furthermore if the original publisher of the article sits at number 4 on Google without the quotation marks why does this sit at number 1:

For the search in quotes, I get a different result set without that page you mentioned, the French page (with the clone) is top. 2&3 are you complaining about Google on two different boards.

At 4 & 7 are scrapers.

You are at 5 & 6.

The solution is for you to complain on various boards and mention some cooking tips. Sort of like:

"Google is all mixed up, but not as mixed up as this tossed green leaf salad with virgin olive oil vinaigrette we made in the Dordogne."

More seriously: It looks like its either a penalty for stolen content (the reviewer has no way of knowing you licensed it), or my pet theory: a baysian duplication analysis done just before the results are returned, used to make the results more 'varied'.

You being an identical duplicate of a higher up page would get wammied down the rankings but you only need to increase your rank a little (assuming that is true) to displace the French site, which would then fall to sub 100.

1milehgh80210




msg:714964
 7:45 am on Jun 13, 2005 (gmt 0)

As a searcher, I've noticed myself using quotation marks more and more over the past year. maybe google needs two buttons, 'search' and 'find'.

larryhatch




msg:714965
 8:00 am on Jun 13, 2005 (gmt 0)

I use quotation marks (or else the exact-phrase box on G & Y advanced search pages)
but I sure don't expect the general public to do that.
I try to rank well with 2 keywords in a certain order, but NOT as an exact phrase.
That would be just asking too much of people who can't even spell 'quotation'. -Larry

ncgimaker




msg:714966
 8:05 am on Jun 13, 2005 (gmt 0)

Hey, I wonder if thats the problem?

Maybe they've just added in a scoring for Bayesian filtering straight into the ranking algo. Like:

Ranking = K1*OffPageScore + K2*PR + K3*OnPageScore - K4* BayesianDupeScore

There's definitely something funky in the algo that has nothing to do with relevence. Perhaps a Bayesian dupe score is it? Since all text in a language is in someway like every other text in that language, that might explain why:

1. I saw foreign language low ranking blogs outranking Google for that Google-Googol-Milton search I did. Arabic would be very different from English, so all English pages would have a slight penalty that the Arabic blog wouldn't suffer.

2. Junk is floating to the top, junk is very different from regular text and so doesn't suffer the penalty because it scores much lower on the penalty.

Suppose they forgot to put a low cutoff mark (or set that low cutoff too low), text about "flange dooberies" would gain a slight penalty for being similar to other pages about "flange dooberies" in the same language causing crud & foreign language texts to float to the top.

ncgimaker




msg:714967
 8:17 am on Jun 13, 2005 (gmt 0)

Ahh of course, that would also cause blogs to float to the surface too.

A blog about 'search' that discussing the rankings for "Cooking holidays in the Dordogne" is very different from normal pages about "Cooking holidays in the Dordogne" and even from "...holidays that will have you cooking in the sun, and in other stories Mr Dordogne...".

Being very different it would score better on the duplication and rise above the rest.

The more I think about it, the more that theory seems to explain the weirdness.

WhoopsAccident




msg:714968
 2:29 am on Jun 16, 2005 (gmt 0)

I have just uploaded the particular page with a new title.

Is it the duplicate title that Google is penalising for or is it the page content?

It occured to me that it might just be such a simple thing as the title being as it was the title that always got my page ranked high.

Must be worth a try.

The other bizarre thing is the name of my web site (which no-one would search for unless they already knew of it). The name is spelt with a hyphen. Spell it correctly (as it appears on the cached page) and I continue to slide (now at 18) but if you type it as two words without the hyphen then it ranks number 2.

WhoopsAccident




msg:714969
 9:30 pm on Jun 16, 2005 (gmt 0)

. . . even more bizarre is the fact that today I am back up there with all my search terms!

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google SEO News and Discussion
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved