Does Google prefer a certain URL format?

and does a page's PR affected by its URL

[flex55]

I'm trying to understand if Google have any preference for PRing pages based on the URL itself- ie, which URL format should I use of:
[domain.com...]
[domain.com...]
[domain.com...]
and so on.

I was trying to see if Google factors URL depth as part of its PR mechanism- meaning, would [domain.com...] get a higher PR than [domain.com...] if all other factors (incoming links etc) were the same?

You can argue that URL path should not be factored in PR, rather the depth of links from the homepage or so. I hoped this is how Google works. But then I ran a small test: I tried to see how Google treeats unknown pages in a known domain name.
Here's what I got:
www.domain.com is pr7
www.domain.com/agets pr6 (even though the reply was 404)
www.domain.com/a/bgets pr5 (even though the reply was 404)
and so on.
Now, these pages are pages that are not in Google index of course (as they don't really exist), so I assume that Google applies some huristics to "guess" the PR of pages it doesn't know in a domain it does know.

What I take from that, is that at least for unknown pages (pages that are not in the index) Google does consider the file path as a factor in the page's PR.

Does anyone know if this applies also for google-indexed pages?

[lazerzubb]

flex55, yes if the pages aren't in the Google index, they guess the PageRank for the pages, based on the indexpage.

I'm not sure, but from the PageRank formula there is nothing saying that you should lose more PageRank just because you are linking to a deper level inside the domain.

And from my own small test this hasn't been the case either.

[Grumpus]

I believe that there are still some dynamically generated pages that ARE in the index, but still don't have a real PR. Most of mine started getting real PR two updates ago, but I'm still finding some that don't make any sense. I think it has to do with the fact that, while Google has been able to "crawl" dynamic pages for a while, it couldn't assign a value because it was having difficulty seeing the links to the pages (or determining if the page was actually different than another page with a different variable in the URL).

That being said, the PR you see for a 404 page is always a Guessed PR and runs at Homepage PR - 1 per "/" or "?" in the URL. I think that the dynamic pages that haven't yet gotten a "real" PR are getting a "guessed" PR assigned to them and within a few months, we'll see that guessed PR going away (except for pages that aren't actually in the index).

Therefore, in MOST cases, the URL doesn't matter, but in a new site or a site that is dynamic and Google is just starting to crawl it, the URL does matter - for the first month or so. Once the bot's gone through and gotten its bearings, though, it's irrelevant.

G.

[lazerzubb]

>>for the first month or so.

Ye i would say that it takes atleast 2 updates, before you start seeing a real PageRank.

[flex55]

Thanks lazerzubb and Grumpus :-)
Very helpfull. I guess I shouldn't care that much about the URL structure if I plan long term..

f.