Google Keyword/Keyphrase Selectivity and Dilution

[Dolemite]

I have what I would consider an SEO opportunity/decision for a new site I'm working on. For a link and filename I'll be using quite frequently, I can potentially include 2 keyphrases in a meaningful and logical way, but I'm wondering if by trying to optimize for both keyphrases, I might be reducing the effectiveness (measured as SERP placement) of both keyphrases.

keyphrase1 = "[type] widgets"
(There will be multiple permutations of keyphrase1 due to several different types, I.E. "furry widgets", "fuzzy widgets", "hairless widgets", etc.)
keyphrase2 = "blue widgets"

As I see it can do any of the following (there are more combinations, but these are what would make sense to the user):

• Optimizing for keyphrase1:
  <a href="/widgets.php?id=[type]">[type] widgets</a>

• Optimizing for keyphrase1 in anchor text and both in filename:
  <a href="/blue-widgets.php?id=[type]">[type] widgets</a>

• Optimizing for both keyphrases in both anchor text and filename:
  <a href="/blue-widgets.php?id=[type]">[type] blue widgets</a>

I want good SERPs for both keyphrase2 and the multiple permutations of keyphrase1, but because of high competition, I'm better off doing well for one or the other as opposed to doing just OK for both. I realize that the weight of the php variables (everything after "?") in filenames is reduced. Unfortunately, I have shared hosting on a server with no mod_rewrite to work with.

Should I be worried about diluting the value of my keywords/phrases if I go with the third option and optimize for both keyphrases in both the filename and anchor text? Will google still pick up on the SEO value of keyphrase1 in the filename even with [type] as a php variable? Will the presence of "blue" in the anchor text phrase "[type] blue widgets" reduce the SERPs for "[type] widgets" (assuming that the search is done without quotes and therefore not on the phrase but the keywords together)?

[ciml]

That's a very good question Dolemite. The problem is that it's not always easy to find sensitive, objective tests for subtle weightings.

I think that the answer is yes, there is some dilution but it's small. It's not something I'd lose sleep over.

<correction>
I do lose sleep over that question, lots of it. I don't loose sleep over the effect in real-world situations though.

[Dolemite]

Hmmm...yeah I wasn't sure there would be specific advice for such a subtle and particular situation, but it seems worth considering and bringing up at least.

I'm sure keyword/keyphrase proximity, order, and context all have an impact. Exactly what that impact is, its hard to say.

I think the best strategy is to make things make the most sense to the user, while still keeping in mind what your keywords/keyphrases are and use them where appropriate and in the places that carry the most value, like headers and anchor text.

For my specific situation, I think I'll be going with a variation of the second option, mostly to keep things logical and succinct to the user, while still getting my second keyphrase in the filename. That should also minimize the lost sleep. ;)