Keyword Density, Stemming, proximity.

[JudgeJeffries]

If I optimise for a phrase ie keyword1 keyword2 keyword3 and that phrase pops up on my page in all the right places does it matter if a number of the individual keywords also appear in the text individualy ie 2 of keyword1, 4 of keyword2 and 1 of keyword3 or is it better to clear the out the individuals to leave the keywords that appear on the page, only appear in the phrase sequence.

[BGumble]

Google prefers when the terms are next to each other. If your page says:

"Come visit us if you're a keyword1 looking for lots of keyword2s that really kick keyword3!"

and your competitor says:

"Come visit us for all of your keyword1 keyword2 keyword3s!"

G would certainly prefer the competitor. The algo uses some sort of "near" analysis to determine how near the terms are to each other which improves relevance. (These are just anecdotal observations.) Having the keywords elsewhere on the page won't hurt you, but its always better to have them together if people typically search for the phrase.

[JudgeJeffries]

I was concerned that the individual use of the keywords on the page may effectively increase the phrase density to an unacceptable level and downgrade position on serps, especially if the SE analyses those individual words that are close.
Does anyone have any idea if its best to remove the individuals

[metablue]

This is kind of related to my question about optimizing for multiple keyphrases. No one seems to know whether optimizing for several keywords dilutes the relevance of each keyword, or whether it's just the number of keywords of each type that matter.

In other words, is there a maximum total relevance or not. If there is, then every time a page increases in relevance for one term it declines for another. Then by having the keywords on their own, you might be hurting the keyphrase.

I think that having the keywords on their own does help, but not as much as keeping them in their phrase. I've noticed that one of my pages is ranked well for a keyphrase that is only mentioned as a phrase once or twice, while one of the words in the phrase is mentioned frequently. It's not because of link anchor text.

You can probably get a higher keyword density by using the individual words, since it could be difficult to keep repeating the whole phrase and still have the text look natural, so that should help your ranking in the end.

[metablue]

Just read message 3 - I guess it depends on the keyword density over all. Anecdotal evidence again - I haven't had any trouble combining keyphrases and the individual words, even when the individual word density is quite high, in fact it helped my rankings a lot. But I just wrote the text in a way that seemed natural. Wasn't trying to stack keywords.

[PatrickDeese]

My rule of thumb for writing keyword rich copy is that if you can read it aloud to someone with a straight face, it is okay.

BAD:

WIDGET WIDGET WIDGET!

Widgets for the widget-minded widgeteer. Widget experts agree, our widgets are the widgets that widget buyers want.

----

BETTER:

The Widget Shop

We offer a broad selection of widgets from the best [manufacturers].

Feel free to browse our [widget catalog] where you can find widgets in any [model], [color], [size] and [price range] you desire.

if you can't find the widget you are looking for we can probably [special order] it.

---

This is pretty short to be a good example, but this formula seems to work for me. :) I never really pay attention to keyword density, at least not in the sense where I have some program analyze it for me. I figure if it is human "tolerable", it will pass the algos.

[JudgeJeffries]

Are you effectively saying that if it passes an auto grammar check, it should be OK as notwithstanding good English on my site, I find that the grammar check frequently kicks in because of my use of lists and also disagreement over fragmented sentences.
Does Google perhaps have a grammar/spell check as part of its algo?

[Mohamed_E]

> The algo uses some sort of "near" analysis to determine how near the terms are to each other which improves relevance. (These are just anecdotal observations.)

Not really anecdotal, here it is straight from the horse's mouth at [www7.scu.edu.au...] in section 4.5.1:

For a multi-word search, the situation is more complicated. Now multiple hit lists must be scanned through at once so that hits occurring close together in a document are weighted higher than hits occurring far apart. The hits from the multiple hit lists are matched up so that nearby hits are matched together. For every matched set of hits, a proximity is computed. The proximity is based on how far apart the hits are in the document (or anchor) but is classified into 10 different value "bins" ranging from a phrase match to "not even close". Counts are computed not only for every type of hit but for every type and proximity. Every type and proximity pair has a type-prox-weight. The counts are converted into count-weights and we take the dot product of the count-weights and the type-prox-weights to compute an IR score.

[JudgeJeffries]

Back to my original question.
Is it perceived as more effective to clear out all the individual words and rely on those words only as a group ie a keyword phrase?
Does anyone here do that in practice?

[FourDegreez]

In my experience, multiple keywords do not dilute. I have a page that ranks on the first page of results for generic-keyword1 and generic-keyword2, as well as for "keyword1 keyword2", "keyword3 keyword2", "keyword4 keyword5 keyword2", etc. Seems to me that Google ranks pages on each keyword or phrase completely independently.

I definitely aim for keyword proximity on the phrases, though.

[martinibuster]

Good post Mohamed_E.

Those pages are so loooong- but definitely worth the read. Those who don't get around to reading it (or rely on WW too much) leave themselves at a disadvantage.

I think that the proximity thing is important to watch, and that paper pretty much makes that explicit as far as degrees of proximity.

[Mohamed_E]

Those pages are so loooong- but definitely worth the read. Those who don't get around to reading it (or rely on WW too much) leave themselves at a disadvantage.

That paper is one of very few that I have printed out, much too long and too important to read on the screen. And far too valuable to scan!

[PatrickDeese]

>> Are you effectively saying that if it passes an auto grammar check, it should be OK as notwithstanding good English on my site, <<

Well, I don't use grammar checks. I suppose I have the "advantage" of getting a BA in English Literature, so the copy writing stuff comes pretty easily to me.

I guess what I am saying that if you feel like the copy is something that flows naturally and doesn't feel like ad copy, you are on the right track.

Obviously the grammar rules for a paragraph aren't going to coincide 100% to the format of a webpage, simply for the reason that web pages are not structured like an essay would be.