Forum Moderators: Robert Charlton & goodroi
This is just a theory from someone probably not as experienced as you!
Even when all your pages in a site go suplimental, the url that is deemed the homepage never seems to.(even when copies of that page i.e. index.htm/ non www etc. seem to)
This homepage could be a copy of 5 other homepages on other sites, but I have yet to see the page deemed as the homepage as suplimental.
This makes me think that a homepage has a different criteria or algorithim to its sub pages and can get away with more.
After all Google may accept that a homepage can mearly be a portal to what is inside the webside and unlikley to hold the real content, just snipets of content at best and contain many internal links.
This is where I think I have made a mistake in SEO over the years, building up links to my hompage where now I believe is deemed less important than its subpages.
My subpages have the content and this is what google likes
This was brought home to me when visiting Matt Cutts site and noticing he does not have a homepage. Just a few words to click here.
No graphics, design or content of any description, why would he do that if there was an importance to the homepage content?
Of course there have been many threads about deep linking to get pages listed in the google index, but I think it goes well beyond that. I believe you should only be linking to internal pages and the homepage will naturaly rise with the rest.
An honest opinion only from a non expert.
Mark
This was brought home to me when visiting Matt Cutts site and noticing he does not have a homepage. Just a few words to click here.
I have two such sites, with homepages left blank other than a mere link to the blog page. I am planning to develop the sites whenever I find time and a good idea! Installing the wordpress into the root would not allow this.
Surprisingly he doesn't include a trailing / on that URL on his internal links.
I think that Matt's site is in part an experiment, encompassing a lot of the vicissitudes of being a webmaster, and this lack of the trailing / could be one manifestation of that.
There's lots of evidence, eg, that Matt was completely aware that he hadn't redirected the non-www form of his default url to the www form (or vice versa... whichever he might pick), and he still hasn't redirected it. I'd bet he's doing a lot of canonical testing.