Google Penalty for path depth

[mmurdock]

OK Google experts - I need to better understand what is going on here.. It appears that google lowers the PR the further down in the path tree your page gets. To test this I set up a static site at www.example.com and get the following pr results with exactly the same content on each url.

URL
www.example.com/ = PR5
www.example.com/default.htm = PR4
www.example.com//default.htm = PR4
www.example.com///default.htm = PR3
www.example.com////default.htm = PR2
www.example.com/////default.htm = PR1

Is the logic behind this assume that no matter what is on the content page the deeper it in in the site the less relavent it must be? This seems to be an arbitrary demotion as I might just as well organize my best content in deeper folders. It also encourages placing all your content in a gigantic (and messy) root folder to get the highest PR?

What gives?

-- Mike

[edited by: heini at 2:25 pm (utc) on April 17, 2003]
[edit reason] examplified / thanks! [/edit]

[BGumble]

Welcome to Webmasterworld!

Had those example pages been indexed by Google or did you just put them online now? If they have not been indexed, Google is simply 'guessing' the PageRank-- they would not receive an actual ranking until the next crawl and update dance.

Google does not determine real PR by directory depth-- it is solely determined by what pages link to that page. To Google, the URL string is just a piece of information, it's not counting the directory levels as that would penalize users on shared servers, etc.

[takagi]

Hi mmurdock, welcome to WebmasterWorld. The path depth is not important to Google. Usually sites get the most links to the homepage, and from the homepage the PR is spread through the site. In some cases a lower page has a higher PR than the homepage.

Use the DMOZ as a good example. Until recently the about page of DMOZ had a PR10, where the homepage was usually PR9. Some high editors have a direct link from their profile to a very low category. This link makes the PR in that category higher than the category one layer higher.

[martinibuster]

This is one of the most elementary and commonly known aspects of PR distribution: It's not directory depth that's responsible, it's how many clicks away from the home page that determines the lessening of the PR.

The reasoning is that the most relevant content is at the top of the page, the least relevant at the bottom: Algo scores are weighted accordingly. Likewise, on a macro scale (site-wide), the more important stuff is closer to the main page, the least important many folders down.

Incidentally, that is why I always cringe when a client wants to have an empty page in front, and make users "drill down" for their information...

If you take a moment and read some of the older posts, you will get a better idea. You can start at the SEM forum and work your way backwards.

This is a great way to get up to speed on what's going on.

Welcome to WW!

[mmurdock]

As heini has edited out the real site url I can't show you the real world example..

What I am trying to determine is why the very same content gets "dinked" in PR simply because it is located at a deeper depth in the URL.

Google has already indexed the content at www.example.com and gives it a PR5. When I access the same content page as www.example.com/default.htm instead I get a PR4. When I access the same content as www.example.com////default.htm I get a PR2.

I must assume therefore that the Google bar PR# is a "guess" and moving the same content page further away from the "home page" causes this "guess" PR to get reduced.

Thanks for the reply..

[ken_b]

hmmm... I'm not sure, but could this...

why the very same content

be part of the problem? Duplicate content?

[mmurdock]

ken_b,

I can't imagine that it has to do with the very same content as there are no links to the www.example.com////default.htm. I simply inserted the additional slashes to see what the google PR did and was surprised at the results. The additionaly slashes are ignored by the web server and the same /default.htm file is served up regardless of the number of slashes.

I will try moving the content down into real sub-folders and see what the PR# does.

-- Mike

[BGumble]

If there are no links to [example.com...] then Google has not really spidered it. It is just guessing at the PR and assuming it is a few directories deep.

[takagi]

Google's PR isn't a ranking for good content. It is about being linked. DMOZ clones have more or less the same content but except for Google Directory and DMOZ itself, the PR is usually extremely low. If you have the same links to

www.example.com/f1.html

www.example.com/dd/f2.html

www.example.com/dd/dd/f3.html

then they will all 3 have the same PR.