Keyword Dilution

(I think this is a MAJOR issue)

[franklin dematto]

This is a MAJOR issue that I am VERY confused about.

Does having multiple keywords on the same page dilute them? That is, if someone searches for "widgets", will a page with the only keyword being "widgets" rank higher with a page that has the same density and position etc for the word "widgets" but also has the keyword "slicing"?

If this is so, this is a MAJOR issue - not only must pages be optimized for keywords, they must be unoptimized for other, insignificant words that may happen to naturally occur frequently on that page. What's more, if there are multiple, alternate keywords that *are* relevant, it would be better to pick one (or at least one phrase), and ELIMINATE the other from the page!!!

(This, in my opinion, is BIG!)

I'm very confused about this, especially, since I thought this goes against the whole idea of themes. I thought that a result of themese is that order to rank high, a page must be optimized not only for the words that are being searched for, but for all relevenat, related words - i.e. that the SE generates a list of relevant words, and checks for that also. Am I totally wrong about this?

If you haven't noticed, I think this is a really key point, and I am very confused, so if anyone knows anything, please, speak up!

[bufferzone]

I don't know for sure, But I try to concentrade on one keyword pr. page. I find it the better way

[SmallTime]

So, If taken to the logical extreme, I eliminate all extraneous possible keywords -I am left with just my keyword. (and stop words) Certainly would save typing:)

If the SE's are looking for natural language, then perhaps it is possible to over analysis the problem.

If I am doing ten pages of widgets, each page will have a little different focus, some of them will have keywords that are a subset of widgets, (i.e. sliced widget) all on the theme of widgets. One page will focus on widgets as a keyword. Not all of these pages will necessarily be "keyworded" or searched for, but do contribute to the theme.
My take on it, but then I may also be confused!

[jeremy goodrich]

This also depends on the type of algorithm you are interested in promoting for...take AV, for example, or the term vector database (do a site search, lot's of cool stuff).

I believe that with a database designed this way, the algo tried to figure out that golden "key word phrase" that the page was about...more modern, and imho better, algorithms index the whole content of the page (AV didn't always do that) and thus, you can get traffic on a whole keyword spread, not just one phrase, per page.

Honestly, when I comb through a referrer log, the thing that stands out is the fact that ten people arrived at the same page from, say, google, while typing in ten different, equally meaning wise, words and phrases...hope these ideas help, just some things to think about.

[franklin dematto]

Another related point is click pop: Is click pop measured per exposure or per keyword? If it's measured per exposure, than you want to make sure that you do NOT appear under irrelevant keywords.

Anyone know how it's measured?

[Liane]

franklin,

I'd sure like to know the answer to that one as well. I have optimized several pages for at least three or 4 keywords and it has worked for me!

I am guessing that it really depends on what your competition is doing. If somebody optimizes for "widgets" alone while his competitor optimizes for "widgets, springs, nuts, bolts" then I would imagine the fellow who is only targeting widgets will ultimately come out ahead on that one search term.

The problem of optimizing for only one phrase per page is ... what exact term is one supposed to use on the index page? I have far too many options to choose just one as THE most important phrase. As a result, I have tried to optimize the top 4 phrases and words and then have optimized the second tier pages for the next most important terms and so on.

Its really difficult to get a handle on it all and I guess it is the very reason that doorway pages were concieved. I don't use doorways ... so as a result, I have several pages which come up in the top 3 for specific keyword searches.

Anyone have any comments about this approach. Am I making a huge mistake doing it this way?

[franklin dematto]

I guess there are two issues:

1) Will having one keyword/phrase take away from the score for another one?

2) Will having keywords associated with one category take away from the score for another category? If yes, how far apart must the categories be?

Can anyone help?

[skibum]

You might try focusing a page on "widgets" with that term appearing more often than anything else on the page and then targeting complementary phrases as well. Widget cutters, widget wackers, widget makers, widget machinery for example. Most of the time that leads to rankings on the complimentary phrases and with enough link popularity the site will start to move up on the single (and most often MUCH more competitive) single word.