• Register
• Log in

SEO Theory - SEO Theory and Analysis Blog

Algorithm analysis, Web community relationship analysis, SEO practices and techniques, industry news, etc.

• Home
• About SEO Theory
• SEO Glossary

Managing PageRank or, ‘Why are so many of my links supplemental?’

Posted by Michael Martinez on July 6, 2007 in SEO Theory

Finding Supplemental

People are looking for a query that shows them all their Supplemental pages. Here’s a clue folks: There is no such query. There never was. If you’re thinking, “But when I subtracted -vbfjg” or whatever, that query did not work. Why is that so hard for people to accept?

Your pages are crawled by both Googlebot and Supplemental Googlebot. Each robot feeds its own index. The odds are very good that, right now, as you read this, one or more of your “non-supplemental” pages are ALSO in the Supplemental Results Index. So, if we assume for the sake of discussion that there is or was some query modifier you could use to see which of your pages are in the Supplemental Index, that won’t tell you which of your pages are only in the Supplemental Index.

Hence, there is no filter supplemental results query. None. Nada.

Let’s put this into perspective.

How many indexes does Google use?

When you type a query into Google’s Web search, Google now looks at a lot of indexes: the Main Web Index, the Blog Index, the News Index, the Video Index, the Image Index, and the Maps Index. AND (maybe) the Supplemental Results (Web) Index. That’s a lot of indexes. And you know what? I suspect that Google has overlap between the Main Web Index and each of those other indexes. I’ve never seen them say so but I think that is the case.

Now, I would love to have a query operator that says: “Please search this index”. There is no such publicly documented query operator of which I am aware. You can use Google’s interface to swap between main index searches but you cannot get to just the Supplemental Results Index, and even if you could you’d still have the problem of what to do about overlap. In fact, you would most likely have that problem with all index-specific queries.

What you really want, and what no one has yet produced, is a query that shows you all the pages that you have in the Supplemental Results Index that are not also in the Main Web Index. Of course, even if you could know that, what good would it do you?

Ah. Now we’re getting down to the theoretical stuff.

Supplemental Google PR Leaking

Some people out there apparently believe (or perhaps merely fear) that you’re leaking (toolbar) PR through your Supplemental pages. Oooo. That’s spooky. It’s also nonsense, as is the entire “leaking PR” concept. But there is a connection between Supplemental Results and (Internal) PageRank.

As we all now know, Google says that every sites gets “only so much (Internal) PageRank” and that you have to spread it across your pages. Of course, a lot of sites have less (Internal) PageRank than they have pages that need it. Hence, their pages “go Supplemental”. More properly, their pages fall out of the Main Web Index.

Rather than thinking of PageRank as a value you capture and store (which was never the case), you should think of PageRank as a ticket for a tour of your Web site. You need to send Googlebot (which follows Internal PageRank) on a virtual tour of your pages so that it will (at least temporarily) include them in the Main Web Index.

Now, here is where the theory comes in: The more often you get Googlebot to enter your Web site through a page that flows to other parts of your site, the more of your site will be included in the Main Web Index. You can actually have earned a lot of PageRank but are distributing it inefficiently (in which case you deserve to Go Supplemental for being an Evil PageRank Hoarder).

Don’t believe me? You don’t have to. I’m getting pages into the Main Web Index across multiple domains with relatively few inbound links. How are you doing?

Conclusion about link building

Link building begins at home. I’ve said that before, and many other people make that point too. But here is where the real SEO Theory comes into play: your external linkage should not be aimed either randomly across your site or expressly at your home page.

Think about it. If your site is large enough that you can have important secondary “index” pages (call them secton headers, or category index pages, or whatever), most of your artificial external linkage should be divided between the root URL of the site and the section headers. Random natural links will splatter across the Web site, but since you’re so devoted to link building you need to focus on where the most important content is.

So each independent section of your site has its own index page. This page links down to all the other pages in that section, and your sectional PageRank flows in the right directions. Except that’s not how it really works. If you’ve been hoarding PageRank you’ve probably been flowing it all toward the root URL in some foolish attempt to boost your Toolbar PR. In other words you’re telling Googlebot you only have one important page on your site.

GoogleBot needs to move around a site, see that the visitor can easily get from one page to the next. The more internal links you have the better, but they should be organized well enough that several things become apparent to the software that Google uses to analyze Web documents:

1. Your content pages are relevant to each other
2. You trust and value your own content
3. You favor (through your lnkage) the truly most important pages on your site

Favoring your most important pages should in some ways validate the inbound linkage you have pointing at those pages. That is, statistically, your internal linking structure should agree with what other people decide are your most important pages.

If you only promote one page on your site, if other people only know about one page on your site, then how important should anyone (including a blind algorithm) regard the rest of your pages to be?

Value begins at home. Value is where you place it. Value is not what other people pass on to you, it’s what you assign to each of your own pages.

Any new Web site can get into the Main Web Index through good crawling and a handful of links from trusted content sites. It doesn’t take a lot of link love to convince Google you have a real Web site worth including in the Main Web Index.

Managing PageRank

You need to manage PageRank through your site structure, not through your links. Let the structure dictate where the PageRank flows because once it stops flowing you no longer have any PageRank. It’s an ephemeral concept that is only useful when Google is calculating PageRank. It doesn’t matter if you see a 2 or an 8 in the Toolbar on any given day. What matters is how often Googlebot hits any given page on your site.

If you don’t have enough PageRank to get the entire ste into the Main Web Index then you need to get all of your most important pages into the Main Web Index, and those most important pages need to link to their most closely related pages with good, descriptive, clearly visible anchor text. You need to tell people both through the pages that are indexed and through your own navigation that you have the content they are looking for.

PageRank Internal Link External Link

PageRank is the effect your pages realize during an event. It’s not a capturable quality or value. The more links you put on your page the less PageRank each link passes on, unless you get more PageRank into your page. If only 1 in 5 of your pages are being included in the Main Web Index, you’re only starting out each PageRank event with 1/5 of your base potential PageRank.

Those internal links are more important than external links because once your pages are included they may or may not pass value to other sites. You may have 50 outbound links on a page but what happens if none of those 50 pages are in the Main Web Index during the next PageRank calculation? I won’t pretend to know the answer to the question, but if you don’t give the algorithm another choice it will make a choice for you: it will reallocate that PageRank across all the rest of the documents in the Main Web Index.

The weaker your internal linkage is, the less likely you’ll include all of your pages in the Main Web Index (you’re counting on external links to get those pages in, but what if your external links are not in the Main Web Index?).

The fewer internal links you provide yourself, the less likely Google is to find and crawl the pages on your site with GoogleBot. GoogleBot is driven by PageRank. Supplememtal Googlebot is not. You need to share your earned PageRank internally as much as possible. PageRank is not going to make or break your search results rankings. It never has. It will, however, make or break your crawling, and that in turn will make or break your search results rankings.

SEO “old domains”

This is one of the oldest PageRank manipulating tricks in the book. There is certainly no reason not to take advantage of the inbound linkage pointing at old domains, but if you’re paying a premium price for those old domains you are paying too much. Spammers will burn through domains in about 30 days. You should make a domain last 30 years. In that 30 years you can rebuild the content completely. Your life won’t be over. I’ve done it. I had to do it. Had no choice. I’ve survived. You can survive too.

Advanced SEO Techniques

The most advanced SEO technique for Google today is the correct management of PageRank. SEOs have led themselves down the garden path with books of PageRank nonsense. Now many SEOs are caught up in a maze they cannot escape because they built the hedges too high. This is what comes of people constantly repeating nonsense on forums and blogs without questioning what they read.

All the PageRank hoarding techniques people have followed — much against my own advice and advice from other “old school” SEOS — are in many ways responsible for a lot of today’s Supplemental Results Pages. After all, when you look at many large content sites where people funneled their linkage toward the root URL, you can see the result of PageRank hoarding. It bit those idiots on the butt and it bit them hard.

Why do pages go Supplemental?

Why do pages go Supplemental? I would say that Google just doesn’t trust them. A well-linked page should be very trusted. That, of course, is why spammers use robots to drop links in forums, blogs, and profile pages. The more links they get to their pages, the more natrural those pages should seem. And the more anchor text they will (theoretically) receive.

So you went out and traded links with lots of sites. No reason not to since you found ancient SEO blogs, tutorials, and forum discussions where the titles spoke about reciprocal linkage.

You went out and bought links, dropped links, submitted articles and press releases, and blogged about yourself because you found plenty of SEO tips that said you should do that.

And those links are pretty much worthless now. Most of them are stuck in the Supplemental Results zone, where they belong. I feel no sympathy for people who follow “conventioanl SEO wisdom” without considering the consequence of doing what everyone else is doing. When a large number of people manipulate links the same way, general, filterable patterns of abuse will emerge. The search engines will be able to put a stop to what people are doing.

Things you should do for SEO

1. Keep your secrets to yourself
2. Find and follow your own path
3. Try one new experiment a month (it doesn’t have to be risky)
4. Ignore all forum discussions and blog posts that casually mention “PR” as if it is useful
5. Ignore all forum discussions and blog posts that talk about backlink counts, link building, Yahoo! Site Explorer, and link popularity
6. Focus on writing copy that emphasizes, truly emphasizes, the concepts you want people to find you for
7. Look at the search results pages as clay that you are helping to shape — only helping, because other hands are molding it too

The best SEO advice is to ignore SEO advice, not because all SEO advice is bad but because no SEO advice is distnguishable from other SEO advice. There are times when I will use Yahoo! Site Explorer. There are things I look for when I glance at backlink reports. There are reasons for why I “build links”.

But there is no value in doing what everyone else is doing because it will never set you apart from the crowd. You may be one of those PageRank hoarders who hasn’t gone Supplemental. Maybe you got lucky. More likely you did something else than just hoard PageRank. If you’re convinced that all you did was hoard PageRank, then you’re overlooking the obvious things, such as the fact that you also created a Web site.

Content rules the search results. That doesn’t mean links don’t play a role in shaping those results, but links work in far more ways than most SEOs give them credit for. The most important links you’ll ever get are the links you create for yourself with your own Web site. The yellow brick road starts right here in Munchkin Land.