Site Directory Structure Effects

What are your thoughts on usings /KW/kw.html in Google?

[webhound]

Hi there

We are in the process of building a new site and are attempting to organize the files into logical locations within the www folder.

What are your comments on structuring things like this:

domain.com/related-keyword/keyword/style-keyword.html

or is it better to go:

domain.com/keyword/style-keyword.html

Or

domain.com/related-keyword/style-keyword.html

I want to organize the files and optimize for better rankings, but don't want to go overboard in the process. Has anyone experienced a better naming convention that provides optimization and easy/logical organization of files?

Note: We have approx 500 pages, consisting of 2 main product groups and then styles within these groups.

Thanks - appreciate any feedback you all have to offer!

Webhound

[webhound]

anyone?

[buckworks]

My opinion: Keep your URLs as short as your system will let you. It makes good sense to use one or two keywords so your URLs are informative to humans, but don't go overboard. Keywords in the URL have a small effect compared to other things you could do so there's little point in long, keyword-stuffed URLs.

[moftary]

and I suggest you to stay away from this strucure. IMHO, google awards sites with natural?x=bluh&y=bluhbluh that those stuffed with keywords. Not to mention that google might penalize your site for this strucure even if it was a searching engine friendly (over optimisation penalty).

[webhound]

yeah thats what i was wondering. ie is using the keywords as folder names over doing it.

I think we'll stick to something short.

domain.com/product-type/products/style.html

This seems logical and natural. No over doing it as thats the last thing we want to do. I am just trying to organize the files, and help with the theme of the site.

Or you think this is still over doing things?

[roodle]

I don't think you'll have any problem with that. I've just revamped a site and revised the directory structure, such that I end up with addresses similar to (in the most extreme case): keyword1keyword2.com/keyword2-keyword3/keyword4-keyword2-someword/someword.aspx?someparam=keyword5&someref=number. Google spidered the entire site and, as far as I can tell, I'm not being penalized for those keywords employed in these kind of addresses.

I can't see how G can penalize you for using a logical structure, albeit keywordy. If that's what your site is about, how can it be wrong to use those words!? I would only use 1 repeat keyword maximum in any address though (excluding domain name)

[bumpski]

If you look at scraper site directory structure you tend to see a lot of keywords and a lot of repetition. Two of three keywords appear in paths repeatedly. Google is doing a pretty good job of making these pages "Supplimental"

I believe Google, in order to eliminate scrapers, as of the Bourbon update, is penalizing keyword heavy directory structures with lots of repetitions of multiple keywords.

Of course this is a killer to product sites and I have one where pages seem to be "URL only" for no apparent reason. But only product pages which used the same multi-keyword directory path repeatedly. This is still just supposition at this point.

Also in my opinion Bourbon seemed to mandate absolute links to all pages. Links can get cumbersome with extensive keywords, but of course they are very discriptive, which can really help a visitor.

[modemmike]

So now I'm curious, is better to have the pages all in the root of the site or break them out into single layer folders?

www.example.com/keyword-keyword.htm
---- or is this better ----
www.example.com/keyword-keyword-keyword/keyword-keyword.htm

The second is more logical in my case and I wouldn't change it because helps organization but just wondering if it's hurting me in the long run?

[roodle]

modemmike, I asked myself the same question a while back, and after a bit of reading around I gathered that it's not the directory structure that matters, but the way the pages are linked up. i.e. a link from the homepage to somepage.htm in the root level should be regarded as equivalent to a link from the homepage to /some-directory/some-other-directory/somepage.htm. I'm still not totally convinced though! It might be purely a PR perspective. Maybe others can comment...

[modemmike]

I have one site that is a PR6 for no good reason (no BL's) but as I drill down (into sub folders) the pages get less PR so I was curious... this site with PR6 is a complete mystery to me PR 6 and no BL's? That's probably another topic however.

[roodle]

The thing is, the Google toolbar is just "guessing" the page PR, based on the homepage PR (I think). I've read that you shouldn't pay it too much attention. I don't think anyone really knows the true PR their pages have.

[bumpski]

modemmike

Would you mind sending a sticky for the URL of the mystery PR6 site. I just looked into another PR 6 and it was interesting.

Certainly Google uses the linking structure of your site to distribute page rank. I maintain a traditional sitemap.htm page with absolute links to most of my pages. Without this sitemap.htm many of my pages seemed to go URL only after Bourbon. I pulled and reinstated this sitemap twice, each time when the absolute links were gone more pages were URL only. The user navigation linkage is "relative" and would have a much deeper struture without the sitemap.htm. The pages that were "deep" were the ones disappearing. The sitemap makes the structure almost flat.