Supplemental index question.

[billshorters]

I read a post by Adam Lasnik from Google in the Google Webmaster Help Group on the supplemental index. He said that the way to bring pages out of the supplemental index into the main index is to "Get more quality backlinks. This is a key way that our algorithms will view your pages as more valuable to retain in our main index".

If you have thousands of product pages from your site in the supplemental index how in practice to you get them all "...more quality backlinks"?

[glengara]

I'm beginning to find the answer to everything is always "more quality links" increasingly boring...

The alternative is to shorten the steps from home page to product page, but that may well be equally unrealistic...

[PhilC]

Adam is correct, but it doesn't need quality IBLs to each page - just to the site.

Each site's linkages are evaluated by Google these days, and the number of pages that are placed in the regular index is determined by the linkages score. The rest are put in the supplemental index. The PageRank of individual pages is used to decide which pages to put in each index - higher PRs in Regular, lower PRs in Supplemental.

So you need more good IBLs to the site, so that the site can have more of its pages in the Regular index. On a side note, because of the nature of internal linkages and PageRank, the more important pages are the ones that will end up in the Regular index, but if you want to make sure that certain pages aren't in Supplemental, channel more PageRank to them internally.

[Kostis Panayotakis]

"higher PRs in Regular, lower PRs in Supplemental"

I do not experience this. It seems like a random thing.

[PhilC]

No, it's not random. When Google determines that some of a site's pages are stored in the supplemental index, because of the site's linkage evaluations, "It's pretty much PageRank that decides if it's in the main web index or the supplemental index." - quote from Matt Cutts.

[glengara]

"It's pretty much PageRank that decides if it's in the main web index or the supplemental index." - quote from Matt Cutts.

That's not quite accurate though IMO, the guy could get a bunch of rubbish links that raises his PR but does nothing for his indexing...

[Webnauts]

In March this year our admin implemented a wrong way our SSL key without me noticing that, and all secure portions of our site have been fully indexed along with our standard site, and now all https pages are in the supplemental results, due to canonicalization problems.

About two months ago, we have taken the necessary measures configuring our .htaccess file to add a second robots.txt, but until now we had no results.

We have on our homepage PR 7 and all other pages PR6.

What do you think about that?

[PhilC]

I'm not sure if you misunderstood what I said, glengara.

Google uses a site's linkage evaluation to determine how many of its pages to have in the Regular index. If they have crawled more than that number of pages, the rest will be put into the Supplemental index. The question is, how do they decide which ones to leave in the Regular index and which ones to put in the Supplemental index? That's what is pretty much decided by PageRank.

I don't understand your question, Webnauts.

[glengara]

*Google uses a site's linkage evaluation to determine how many of its pages to have in the Regular index.*

No argument there, but that evaluation is not simply based on PR, IIRC all those sites that dropped post BD due to their (mostly) reciprocal links being devalued suffered no loss in PR.

I just found MCs "It's pretty much PageRank.." overly simplistic :-)

[PhilC]

I didn't mean that linkage evaluations are PR-based. The use of PR comes after that - after they have decided how many pages to have in the regular index. They use each page's PR at that point to decide which ones to have in the regular index and which to put in supplemental.

[Webnauts]

To my previous post above, I would like to say that I managed to get out of the supplemental results. :)

[ddwebguru]

Interesting topic.... i have some experience
1] when google understand the body content is duplicate (i think this the most basic idea)
2] if canonical issue not solved, google will treat the domain name > http://www and http:// as two pages.
3] if title and description are same for more than one page


Seo India (http://www.webguru-india.com/seo-india.php)

[mtb167]

I'm beginning to find the answer to everything is always "more quality links" increasingly boring...

The alternative is to shorten the steps from home page to product page, but that may well be equally unrealistic...

Backlinks may not be your problem at all.

Supplemental results often:


Are orphaned pages, meaning they are only ever linked to once
have duplicate title and description tags, meaning google doesn't take the time to figure out whether or not they're all the same
Are otherwise duplicate content pages.


I find that normally, if your site is 265,000 pages or whatever, the second bullet point is generally the problem.