Natural URL depths

Hi,

I've been reading up on the area of natural urls for SEO and found that MSN and Yahoo tend to devalue links the more directories they're under.

So for example
/store/housewares/kitchen/cooking/pans/copper/ would be devalued in that case.

Is this myth or fact and does google have anything like this?
If so, what is a reasonable maximum depth?

Thanks

From what I've seen Google doesn't take the number of subfolders into account. Of greater importance is the number of links it takes to reach the page from the homepage, hence the reason deep links to your site from other sites can greatly improve your position for deeper pages.

Personally I try to avoid creating too many subfolders within a site as it makes any changes to the structure of your site that much more difficult with numerous pages to 301 redirect.

For example, if the existing URL was /store/housewares/kitchen/cooking/pans/product.html and my product list grew enough to warrant creating a seperate category for copper pans as per your example, all my copper pans would require a new URL and a 301 redirect as a result. If my URLs were set up as /store/product.html I could happily create and delete categories as required and move the products without ever changing the product URL. A good breadcrumb trail would then still help the site visitor locate their position within the store.

JagNet,

I have a question for you here :

Is it that when google follows a link on same domain, it follows it as absolute one rather than a relative one. Because if it follows it as an absolute one, then the character limit of the url may well become a problem in high depth links. Isn't it?

It would have to be a very long URL before this became a problem - 2083 characters in Internet Explorer if memory serves me correctly.

Do you confirm that Google follows it as "absolute" url?

I assume that Google converts relative URLs to absolute ones is some way. How, or when, in the process between crawling and indexing I have no idea -- that would require a Google engineer to answer that.

Personally I always use absolute URLs on a page as a precautionary measure against scraper sites. I make the process easier on myself by writing them relative to the domain root (saves time), then use a simple server-side script to convert them to absolute URLs prior to outputting the page.