Common words and related words in WMT... interesting clues?

[MrFewkes]

"Below are the most common keywords Google found when crawling your site. These should reflect the subject matter of your site."

Now then - I noticed that a lot of words were kind of related more to an ecommerce site than my subject matter. So lets say my subject matter is two words (which it is) - blue widget.

The top two words in the list on WMT are indeed blue and widget - then there are 3 related words (but I hasten to add only vaguely related) then the rest are "add" "cart" "product" "review" "account" "catalogue"

That sort of thing.

Can anyone advise the following :-

1. Where on WMT or anywhere else on google (it has to be google generated) might I find words which are related to "blue widget" which I can put on my page so that the page hits the statement "These should reflect the subject matter of your site."

2. Will loading my page with these words and ensuring they do indeed reflect the subject matter (so that this WMT list I am looking at is filled with great words from google somewhere) actually help me ranking wise for the term "blue widget"

I guess this is kind of LSI - if not TOTAL LSI related.

Any help appreciated.

MrFewkes.

[jimbeetle]

1. You can try the related operator in a Google search [ ~keyword ]. It doesn't appear to work as well -- or at all -- on multiple term searches [ ~"one two three" ], but playing around with it might turn up something useful. Note that autosuggest always includes the related search at the bottom of the box so you have to actively choose it to see the results.

Love you Google. Put what the user is specifically searching for as the last choice. Yes, everything you do is with the user in mind.

2. Loading your pages with these words won't help and might hurt. The best bet is to write naturally and not force anything. Write for the visitor who is going to read the copy, don't let it appear to be stilted. Every site has a different audience; write for that audience in a voice it would expect.

And yeah, it's kind of LSI related, though with everything that's been written about it over the years, nobody really know what type of system Google actually uses. The simplest way to put one of the most popular versions is that if G sees "this phrase" it *might* also expect to see "that term."

Whenever you talk about this "latent" stuff the emphasis is always on the mights and maybes.

[aristotle]

It can sometimes help a page rank higher if you expand its content. If you do some research on the subject and find some new material to add to the page, you might be able to add some new related words this way.

[MrFewkes]

If I did the following two pages about Albert Einstein.

Page1
Title Albert Einstein
Body - contained the following
"Albert Einstein is noted for his general theory of relativity and his work on splitting the atom which gave rise in part to the development of the atom bomb. His most famous equation is e=mc2 and is a measure of the equivalence between mass and energy. His theory of special relativity deals with the speed of light primarily."

Page2
Title Albert Einstein
Body
"Albert Einstein watches television in heaven. He enjoys drinking in his local bar and is notorious for being a drunken slob. He likes to feed his tropical fish and watch them as they swim around the fishtank looking for food. He does not enjoy cleaning out his fishtank though."

Then - each page gets 1 link - and each of the links has an anchor text of "click here".

Each of the links is 100% equal in every aspect - trust rank - PR - content of page on which each link is on is identical etc etc

Each link is located on by google at exactly the same time.

Each page (page1 and page2) sit on the same domain and are orphans.

Which page would rank for the search "Albert Einstein"

[edited by: MrFewkes at 11:31 am (utc) on Dec 4, 2010]

[MrFewkes]

I think the first one. But I have no proof.