site is the same as site's

[janeth]

This was brought up by ronniethedodger.

We wanted to see if Google saw site's as site.

So he wrote the text

"We Can Help You Suceed on-Line By Optimizing Your Web Site's Design And Content For Major Search Engines And Directories!"

The word we where looking at was "site's" we wanted to know how Google saw it.

Last night Ron emailed me and said if you go to

custom site design

Ranked #1 ( unless it moved )look in the cached and you will see that Google does see site's as site.

[janeth]

So this means to me that if you target the Plural you get the singular free.

Is this right or wrong?

[janeth]

It could work both ways if you go to


Optimizing Your Web Site's Design


#4


They count site as site's

So maybe it does not matter to Google either way

[ronniethedodger]

So this means to me that if you target the Plural you get the singular free.

Is this right or wrong?

Wrong. If you target the "possessive" form of the word it regards it as the "singular" also. Somehow the apostrophe is acting like a delimiter, just like commas would.

If you look at he cached version of the page in your case, when one of the words was "site", Google highlights the entire word "site's" (if you have highlighting turned on).

In the case of the "plural" form of the word (sites), it will be regarded as a different word.

[janeth]

So then the "possessive" ends the word there

that means we could do a search for 's and get a lot of sites that use that key word?

So the 's does not help anything?

[cbp]

So this means to me that if you target the Plural you get the singular free.

Is this right or wrong?

Yes. Google has been 'stemming' for a few months now.

See this example (http://www.searchenginewatch.com/searchday/article.php/3286181) from Danny Sullivans site (scroll half way down page to question).

Also this blog entry (http://google.blogspace.com/archives/001092).


CBP

[ronniethedodger]

Yes. Google has been 'stemming' for a few months now.


You know cbp...the more I look at it, the more it looks like this may be one of the key reasons for knocking those two-word keyphrases out of the results.

Think at it from this angle. People have been optimizing their pages for a certain phrase. Repeating it over and over. Sure...they may have it well represented, but you keep hearing the complaint "The number one result only mentions it once" or "heck only one of the words appears on that page".

Maybe stemming is allowing Google to interpret the page their way. By picking up on variations of words, possibly synonyms also.

They know the differences between nouns and verbs, and they can distinguish between the simpler tense forms of words (s, ed, ing). They also are getting pretty good at recognizing naturally written text.

A lot of the 2-word terms often contain everyday common words too. Gift Baskets, Web Design, Travel Agents, etc. I wonder how a word like "goose" would of been affected, or a not so common word like "flugalhorn".

In order for this to be simplistic enough to serve up results quickly, how would it be done? Who knows and who cares, because it is being done right now.

Look at the search for "promotional products" for instance. The number 3 result for Norwood has the word "promotions" highlighted (not the word "promotional"). Two more results below that, Branders, you see this again.

Perhaps the answer is that they are putting more weight on content that contain variations of the words themselves. These words would have to appear naturally and syntatically correct though.

[Mel]

I don't think that Googles stemming is clearly understood at the moment, but the easiest way to check how they are stemming a particular search phrase is to turn on the highlighting in the google toolbar, which will highlight all the words in the search term (and bold all that are relevant) and then do a search for the terms you are interested in.

If you search for website design versus website designs you will see that the search results are quite different and that the word design is not highlighted in the ...designs search but the word design is bolded, which to me indicates that it is regarded as relevant to the query.
If you search for website's design you will see that the word design is also bolded, so I would draw the opposite conclusion, that it is somehow considered as relevant to that search term.

[janeth]

I'm thinking it is what Ronnie started off saying.

That Google ignores the 's

[Mel]

Actually the search results are quite different for the search terms websites design and the terms website's design, although Google does appear to recognize the term website is related to both websites and website's.

If you want to rank well for one or the other you will have to use it in the normal way, but it might be a good idea to make sure you have all three in various places.

[Dragonsi]

Does this include "website designer" & "website designer's"??

[janeth]

Does this include "website designer" & "website designer's"??



Yes I would thank so

[ronniethedodger]

If you want to rank well for one or the other you will have to use it in the normal way, but it might be a good idea to make sure you have all three in various places.

That was my main point I was trying to make, Mel.

But...I also wanted to stress that when you do write for the other forms, then it should also be syntatically correct in it's usage. The SE's are getting more sophisticated in text recognition.

I am also thinking that having the words in various forms sprinkled throughout the page, sometimes together - sometimes alone, has a better effect than placing the two-word combo together repeatedly throughout. It may be that you only need to mention it once, then the page is analyzed for content to support that combination - and ignores multiple occurences of the combo itself. It may even go as far as dinging you with minus points and consider it spamming, especially if they are used out of context or are not syntatically correct.

[Mel]

Janeth the easy way to check is just do a search for each term and note the differences in the search results, and in this case the results are much different for ... designers and ...designer's.

In the search for website designers both website and websites are bolded as are design and designers.

In the search for website designer's google suggests that perhaps you have spelled designer's wrong but bolds both website designers and website designer's although the results for each search are different. I interpret this to mean that google does recognize designers as a root of designer's, but ranks one more strongly than the other.

BTW look at the #9 result for website designer's - its a bridal site which does not have that term in the head or on the page, but only in inbound links.

Ron I would have to see some evidence that Google is ranking sites based on Grammer before I am going to spend much time trying to optimze for syntax.

[janeth]

I still think you get more for inbound links then anything you can do to the web site.

[cbp]

Anchor text beats PR

The paper I referenced in another thread, specualtes that 20% of ranking is on-page stuff - the other 80% comes form off-page.

CBP

[Mel]

Anchor text is a powerful ranking tool in Google, but I am seeing some indications that external anchor text is now much more important than internal links between pages on your site.

I personally never concern my self about PR when it comes to analyzing the reasons a page is ranking well.

There are even those who speculate that PR is no longer used by Google as a part of its ranking algo, many of them pointing to the upcoming Google IPO and the fact that the patent for PR is not owned by Google but by Stanford University.

[cbp]

many of them pointing to the upcoming Google IPO and the fact that the patent for PR is not owned by Google but by Stanford University.


Yes - come across those arguments before. Some try to argue that Google has to get rid of PR before the IPO so they are 'unemcumbered' (or whatever the word is) -- I do not not believe this yet. Obviously Stanford licenced the PR patent to Google and still have an interest in it, but we have no idea what the terms of that license are. It is highly likely that there are no problems at all with the patent and the IPO.

This theory was posted in one place (I think at JimsWorld, but I stand to be corrected) as pure speculation (this was acknowledged by the author) - I am surprised how often it comes up as a fact elsewhere.

CBP

[ronniethedodger]

Ron I would have to see some evidence that Google is ranking sites based on Grammer before I am going to spend much time trying to optimze for syntax.

The fact that Google has the word "promotions" displayed in bold in the results for the query "promotional products" shows that they are looking at other forms of the words in the query. It is also showing that they are using those other forms to base those results on.

How much weight they apply to this on-page factor, I cannot say. But the fact is that they do look at it, and it might mean the difference of being on page one or page two. I will take the page one.

I think the variations on words go a lot farther now, than repeating the same form two or three times. And in order for you to use them in the other variations, then you also have to be syntatically correct in that usage....otherwise it will look like crap when a human reads it.

[minstrel]

This theory was posted in one place (I think at JimsWorld, but I stand to be corrected) as pure speculation (this was acknowledged by the author) - I am surprised how often it comes up as a fact elsewhere.
To me, this is one of the fascinating aspects of the psychology of the internet -

Basically, someone, thinking out loud, says "what if?" and the next thing you know it pops up in 10 other places as "did you hear that..."

[Mel]

Yes Google has finally started doing some limited stemming, but there seems to me to be no evidence that the stemmed variants have to be used on a page to increase ranking, and/or that they have to be used in a synatically correct manner.

[ronniethedodger]

Yes Google has finally started doing some limited stemming, but there seems to me to be no evidence that the stemmed variants have to be used on a page to increase ranking, and/or that they have to be used in a synatically correct manner.

I am seeing more and more of it. Here is an example which I borrowed from your site (hope you don't mind) for a search at Google for economical search engine optimization (http://www.google.com/search?hl=en&lr=&ie=UTF-8&oe=UTF-8&safe=off&c2coff=1&q=economical+search+engine+optimization&btnG=Google+Search). The search is done without any modifiers such as the + sign.

The plural form of engine is picked up in the results. But that is an easy one. There is also the form of the word optimization that is also picked up. The word variation for that was optimize. There are a couple other variations if you look thru the list.

Now in each case of course they are syntatically correct. But, you are right -- that does not mean that Google is checking for grammar. But good grammar is good writing, and is better read by your visitors.

This may also be "limited" stemming as you put it. But it also may enough to move you up a couple of notches too....or even onto page one, eh? ;0)