PageRank Sculpting for SEOs who believe in PageRank Sculpting

So you’re reading lots of blogs that talk about PageRank sculpting and you feel like there must be something to it. After all, so many SEOs are talking about it. And it’s not like the SEO community has ever been wrong about anything, is it?

Let’s take a look at some of the arguments people have made for PageRank sculpting:

1. Googlers like Matt Cutts say you can use nofollow on internal links
2. “Unimportant” pages tend to rank above more important pages because they have too much PageRank
3. You can funnel more PageRank through a single link than through several (on the same page)
4. You get more pages indexed by using “rel=’nofollow’”
5. Everyone quotes Rand Fishkin
6. Altrec used nofollow with great success!
7. Privacy policies and terms of service pages don’t need to be searchable

Most of the arguments I have found in favor of using “rel=’nofollow’” on internal links for PageRank sculpting reiterate these points. I have found nothing else but there may be other attempts at justification I haven’t seen yet. So let’s see how well these reasons stack up against the verifiable facts.

Googlers like Matt Cutts say you can use nofollow on internal links – Actually, a lot of people have quoted Rand Fishkin’s deliberate mis-statement of what Matt Cutts said without quoting Matt’s following comment.

Does Google recommend the use of nofollow internally as a positive method for controlling the flow of internal link love?
A) Yes – webmasters can feel free to use nofollow internally to help tell Googlebot which pages they want to receive link juice from other pages
(Matt’s precise words were: The nofollow attribute is just a mechanism that gives webmasters the ability to modify PageRank flow at link-level granularity. Plenty of other mechanisms would also work (e.g. a link through a page that is robot.txt’ed out), but nofollow on individual links is simpler for some folks to use. There’s no stigma to using nofollow, even on your own internal links; for Google, nofollow’ed links are dropped out of our link graph; we don’t even use such links for discovery. By the way, the nofollow meta tag does that same thing, but at a page level.)

Notice how Rand asked a leading question and he provided his own interpretation of what Matt said first. Matt Cutts came along to rebut Rand’s misinterpretation but no one on the pro-PageRank Sculpting team wants to give him credit.

I think saying people “should be” using nofollow is a bit strong. More like people can use it for internal links if they’re power-user-y enough to want to sculpt PageRank flow within their site at the link level. But I’d say that most regular webmasters don’t need to worry about link-level PageRank flow within their site. I think saying “power users and webmasters should be employing on their sites” overstates it a little. It’s available if you want to get into that much fine-grained control.

So, yes, Matt said you can use nofollow on internal links but advises against doing so. And that was not the only time a Googler has advised people not to use nofollow on internal links (Adam Lasnik advised people not to fuss over it, for example). Nonetheless, Rand Fishkin and the SEO community have turned Matt’s stance on its ear and used part of his words to support their arguments.

At this time, Google’s policy on nofollow does not include any recommendations that people attempt to sculpt PageRank or manage internal PageRank flow with nofollow (except for very special internal content). Google does from time to time change its policy statements but I would be surprised to see them actually recommend to Webmasters in general that using nofollow on internal links for sculpting PageRank is a good idea.

“Unimportant” pages tend to rank above more important pages because they have too much PageRank – People often argue that unimportant pages like “terms of service” and “privacy policy” pages shouldn’t be ranking highly in search results. These types of pages usually rank well for your name because they use it a great deal more often than your other content (that’s an aspect of search engine optimization that pro-PageRank Sculptors should give more attention).

If your “About Us” page ranks above your root URL for your name, you have three options:

1. Use your name more often on your root URL in crawable, indexable, clearly visible text
2. Use your name less often on your “About Us” page
3. Use your name in your internal link anchor text for your root URL instead of “home”

Okay, you could put “rel=’nofollow’” in the links pointing to your “unimportant” pages but the people who claim those pages are unimportant seem to be utterly incapable of performing the most minimal research on whether people search for ‘privacy policy’ or terms of service. If seeing those graphs on Google Trends doesn’t persuade you that people look for this kind of content, check your own server logs (and keep in mind that it takes a LOT of traffic to get a query into Google Trends).

If these pages are so unimportant, why are so many people looking for them? Don’t believe me, just play with the Google AdWords keyword tool and see for yourself if you can reproduce query data like this:

You will find similar results for “terms of service” and “privacy policy”, where people combine those expressions with company and service names.

And you’d better believe that U.S. courts may require you to make your ‘unimportant’ pages available through site search because site search is often the number 1 navigation tool people use (on large sites). If you want to enforce your terms of service you had better well make sure people can find them.

You can funnel more PageRank through a single link than through several (on the same page) – Well, Matt Cutts has said in the past that I’m wrong about how Google allocates PageRank. Fair enough, but I’ve yet to see any Googler or PageRank chasing academic conclude that if you have five links to the same destination on one page that you’ll improve the PageRank flow to that destination by removing four of them. All the technical papers I’ve read say that all links to the same destination are treated as one link before any PageRank distribution occurs.

So what about people who claim to have actually seen the benefit of nofollowing superfluous links? Well, what did they do to account for other possible factors? The SEO community leaps to unsupportable conclusions on a daily basis, especially when Google is churning its results. In fact, most “theories” or explanations for Google behavior are proposed at the worst possible time — when Google is rolling out a new index (data push) or algorithm adjustment (update). People concoct all sorts of bizarre ideas, share them on blogs and forums, and never bother to go back after Google’s search results settle down to say, “Never mind — I was just being stupid and panicky”.

To date, no one has published any credible evidence showing that using “rel=’nofollow’” on internal links has in any way benefitted a Web site’s search rankings. Knocking a more relevant page out of the search results by telling the search engines to ignore it is BAD SEO because that just means you don’t know how to make the right page the most relevant one.

Which leads us to…

You get more pages indexed by using “rel=’nofollow’” – Some people now qualify this by saying “you don’t actually get more pages indexed, just more important pages”. These are the Peanut Butter SEOs who take Matt Cutts’ peanut butter analogy (you get only so much PageRank for your site, so you have to spread it as thinly as possible like spreading peanut butter across a slice of bread) to heart. Their assumption is that the “more important” pages that suffer in the search results do so because the “less important” pages have too much PageRank.

It is entirely credible to say that an “About Us” page could rank above a product page (where the product name uses the company name) if the product is mentioned on the “About Us” AND if the product page is in Google’s Supplemental Results Index AND if the “About Us” page is in Google’s Main Web Index then, yes, the “About Us” page should rank above the product page.

Of course, the “About Us” page could just link to the product page with appropriate link anchor text. After all, if the “About Us” page really does have all that PageRank, the easiest thing to do is share the PageRank rather than choke it off. Not that either solution solves the real problem: if the product page is so vastly important, why did it go Supplemental in the first place? That’s a bad linking strategy (and poor SEO) at work.

Fix the problem by building a sound navigational system and linking profile. Don’t cover it up by using “rel=’nofollow’” as a self-imposed search engine penalty.

Everyone quotes Rand Fishkin – Rand is a smart fellow and very successful. Despite the criticisms I’ve thrown his way I have a lot of respect for what he has accomplished. Nonetheless, I’ve said it before and I’ll say it again: stupidity is the most widespread disease in the SEO community. People will believe just about anything if they see it repeated often enough. Psychologists (especially Nazi psychologists) learned through brutal studies that if you repeat a lie often enough it will become believable to many people.

Well, that doesn’t just apply to lies. It works just as well for bad advice, stupid ideas, and completely moronic behavior. Science has proven that significant marijuana use damages brain functionality, causes long-term addiction, and may be more carcinogenic than natural tobacco. Nonetheless, plenty of people continue to use marijuana and even defend it with pseudo-scientific arguments to justify their use of it.

One might as well say a daily dose of engine coolant is good for dogs. They reportedly love to drink it. It also kills them, but, hey, they keep coming back for more. SEOs are the same way with bad advice and search engine optimization mythology. They love buying into nonsense and they keep coming back for more because they seem to have left their cynicism in their other suits.

“It must be true because everyone keeps quoting Rand Fishkin” is about as stupid a reason to believe something as anything (apologies to Rand — we’ve all been wrong about various things and Rand has no monopoly on that any more than I am immune to it).

Altrec used ‘nofollow’ with great success! – I don’t have time to do a full case study on Altrec. I seriously doubt the quality of the one Altrec case study would bear up well under close scrutiny. If you search on “Altrec” in Google today (tomorrow’s search results may change), you may see Google sitelinks for “men’s jackets” and “women’s jackets”. Clearly, Google feels these are important pages for Altrec. However, when I search on those terms I don’t find any of Altrec’s “important” pages in the top ten search results.

I do find pages from JC Penney, who don’t have Google sitelinks (in my results — your results may differ). It’s hard for me to conclude that Altrec has done anything brilliant with search engine optimization. I also tried queries for “snowboard” and “north face” (which also appeared in Altrec’s sitelinks). No Altrec in the SERPs.

Now, when I search for “North Face boy’s denali jacket” Altrec comes up first — but, gee, LOOK AT THE TITLE TAG: “The North Face Boy’s Denali Jacket – FREE SHIPPING at Altrec.com”.

Look at the meta description: “The best source for buying the The North Face Boy’s Denali Jacket. Altrec.com offers free shipping on The North Face and a 100% satisfaction guarantee”.

Look at the CUSTOMER COMMENTS on the page where people mention “boy’s jacket”, “denali”, “North Face”, etc. Hm. Does anyone other than me see a pattern here? Do you think the page might be RELEVANT to the keywords?

Since I grabbed the boy’s jacket keyword from the front page of the Altrec site search results, I figured I’d grab another. I went with “patagonia men’s dawn shoe”. Once again Altrec ranks first for the keyword. Could this be solid proof that “rel=’nofollow’” did the trick? There are far fewer comments on this page (according to Google’s cache image). Nonetheless, Google’s exact find query for “patagonia men’s dawn shoe” indicates there isn’t a whole lot of competition for the expression.

In fact, an inanchor query shows that barely a handful of sites are attracting “patagonia men’s dawn shoe” links with that anchor text.

This is hardly evidence for a “rel=’nofollow’”/PageRank Sculpting slam dunk.

I feel like Jerry Maguire and Clara Peller combined: Show me the money! Where’s the beef?!

So let’s move on from the unimpressive Altrec random audit.

Privacy policies and terms of service pages don’t need to be searchable – See previous points above.

Have I convinced the SEO community to let this nonsense fade into the woodwork? I seriously doubt it. Based on past experience — and I do not exaggerate — it will be sometime in 2010 before most SEOs are laughing PageRank sculptors off their blogs and forums. It takes about 2 years for any bad idea to fall out of the SEO “collective wisdom”. Why? Because this crap gets repeated and applauded so much that most people just have to end up seeing for themselves that it isn’t worth all the hoopla.

Dan Theiss was probably the first big name to advocate PageRank Sculpting and using “rel=’nofollow’”. When I (and other people like Shari Thurow) criticized his argument he quickly amended his position to be less convinced that “rel=’nofollow’” is the great idea he originally proposed. In fact, he has gone out of his way to point out it’s not a silver bullet, though he still advocates its use.

Rand Fishkin jumped all over Matt Cutts’ comment and pushed it to the max, and I believe it was Rand’s inappropriate restatement of Matt’s point of view that led to all the confusion and misinformation that have since pervaded the community. He went for the link bait, or maybe he really believes he’ll ascend to heaven if he drinks the Kool-aid — I don’t know. All I know for sure is that a lot of people quote Rand Fishkin and not Dan Theiss or Shari Thurow.

We can take away a few things from the whole PageRank Sculpting fad. First, if you have a competitor who is using nofollow on his own pages you can rejoice: you’re racing against an idiot.

Second, people who sculpt PageRank MUST wait to see what happens. They typically don’t know how to isolate factors in their nofollow testing so they’ll misattribute any perceived success to their new tactic.

Third, in order to compensate for their bad onsite navigation, these people will have to work even harder to obtain more links because, frankly, when you tell the search engines to stop crawling your most heavily linked pages, you’re basically telling them not to come back to your site for a LONG time. You diminish the power of your own internal links with “rel=’nofollow’”. They drive most of your crawling, pass most of your anchor text, and are in a better position to give you PageRank where you need it than someone else’s page.

You never know when all those great “high quality” links you engineered from other sites will be nofollowed because their in-house SEOs concluded that it would be a good idea to “sculpt PageRank” or “hoard PageRank”.

In fact, there is no difference between sculpting PageRank and hoarding PageRank. You cannot do it. Dan Theiss cannot do it. Rand Fishkin cannot do it.

Matt Cutts on the best day of his life could not sculpt PageRank without looking at internal data that only Google possesses and never shares.

No one can sculpt PageRank because we cannot see it, we cannot measure it, and we don’t decide which links are permitted to pass PageRank (inside Google’s database) and we don’t decide which pages are permitted to receive it (inside Google’s database). Everything we do as SEOs and Webmasters begins and ends with the Web pages we create and modify. Google and the other search engines take over from there and we cannot do anything about what happens inside their data centers.