Forum Moderators: open
All 80 pages are reachable from any page on my site.
This is already the best way for a flat distribution(*). However, this doesn't mean that the internal pages have as much PR as the home page.
A rough calculation yields the following relation: PR_home / PR_internal approx (N-1)*(1-d)/d, where N is the number of pages (N >> 1) and d is the damping factor. PR is the real PR and it was assumed that PR >> 1.
If you want to distribute PR more equally, I would get external links to different internal pages. However, as long as the number and quality of incoming links (and the number of pages on your site) doesn't change, a total flat distribution would lead to a PR4 for all pages (at least in your case). However, the PR for the internal pages would be increased (even if they would be still a PR4).
_______________________________________
(*) Of course, there are linking structures which would lead to a more flat distribution. For example, you could remove all links to the home page from your internal pages. However, this doesn't seem to be a pratical solution.
sounds like a nightmare for user navigation
Depends on how it's accomplished. I visited a great site the other day which had perhaps 50 pages all accesible from everywhere. Navigation was in the form of drop down menu's - rather like menu items in standard windows apps.
Very easy to find my way around.
TJ
As I see it, to have a flat 5 PR is when your home page is a high five, right on the cusp of a six, AND you have some inbound deep links circulating within your inner pages.
The advantage is the same advantage that higher PR usually confers: If you had a choice, would you want your site's inner pages to be a 4 or a 5?
in general there is no advantage of a flat distribution. It depends on the site: If you want to optimize PR for a few pages of a large site, I would chose a hierarchical structure. However, if you have a large number of important content pages, I would chose a flat structure. The total PR of the site (i.e. the sum of all pages) is always the same.