My personal opinion, based on experience we’ve had working with client sites, is that it can still be important to optimize PageRank flow through a site. In this article, I’ll explain why, along with recommendations on how to do it. And this isn’t just about achieving higher rankings. I’ll also suggest ways to improve your site’s relevance to searchers once they arrive.

The rules have changed

In 2005, largely in response to the prevalence of blog comment spam, Google said that NoFollowed links “won’t get any credit when we rank websites in our search results.” That meant you could insert NoFollow tags as a way to shape how SEO credit flows through a site—a practice often referred to as “PageRank sculpting.” If you tagged links as NoFollowed, all the credit they would have otherwise received would be redistributed to other links, thereby giving those other links additional credit.

Then, at SMX Advanced 2009, the rules changed again. As explained in Matt Cutts’ post, Google would no longer allow credit to simply flow past NoFollowed links. Now that credit just evaporates. It no longer flows to pages that would have received it as extra credit under the 2005 rules. This means that website owners must now use other methods to affect credit flow—such as changing the site map.

Is PageRank sculpting now useless?

In my opinion, PageRank sculpting still has tremendous value. Yes, website owners must find a new way to sculpt PageRank—but with the opportunity to increase website usability and also achieve better SEO results at the same time.

That opportunity comes from exploiting the difference between PageRank sculpting using NoFollow tags and approaching PageRank sculpting as a general strategy.

PageRank sculpting as a strategy

PageRank sculpting in general has and will continue to be an important aspect of onsite search engine optimization despite Google’s change in its NoFollow policy. Consider the meaning of “PageRank sculpting.” Although I have not been able to find a universal definition, here is one I propose, based on the needs of website owners:

PageRank sculpting is the application of rules regarding the quantity, prioritization and relevance of links on pages throughout a site to efficiently distribute PageRank. For example, suppose a page on my site has 100 PageRank “points” to pass on to the pages that it links to. I will want to be strategic about how those points are distributed rather than squander them indiscriminately.

If the page links to too many pages or has less important links above more important links, or has links that are not relevant given the purpose of that page, I will want to make some major changes. My initial aim may be to “sculpt PageRank” in order to improve SEO rankings. However, in the process of organizing my links for better SEO results I should also be organizing my site to make it more useful and more intuitive for users. In fact, it’s almost impossible to do one without doing the other. That’s why PageRank sculpting should be tackled in tandem with the pursuit of good information architecture. Choices that serve the purposes of one serve the purposes of both.

What works for Google works for users

As search engines continue to evolve there will be increasing overlap between the things you do to improve Google rankings and the things you do to improve the experience of searchers who find your site in Google results. Here is a basic example of how PageRank sculpting and good information architecture can work together.

This is our made-up page called “Page 0.” As you can see, it links to 35 pages:

Page 0 has 70 points of PageRank to share by virtue of the other webpages that link to it (not illustrated). Using the classic definition of PageRank that Matt Cutts uses in his blog post, each of the 35 pages it links to receives two points of PageRank.

From an SEO perspective this works if pages 1-35 each have target keyword phrases of equal importance and value. From an information architecture (IA) perspective this works if pages 1-35 all have content highly relevant to the searchers who reach Page 0. But we’re not living in a dream world.

Regarding SEO: Pages 1-35 will be targeting a diversity of keywords. Those keywords will each have varying “potential” based on how popular they are, how competitive they are and how relevant they are to the products offered on the site. We want to sculpt PageRank so that more points go to pages that target high potential search phrases and away from pages that target low or lower potential search phrases. This requires reorganizing the sitemap such that links are selectively removed from Page 0.

Regarding information architecture: Pages 1-35 will have content suited to a wide variety of searchers, but it’s likely that only a small subset of those pages will be highly relevant and valuable to searchers who enter on Page 0. We want to architect Page 0 with the goal of prioritizing highly relevant links, and removing links that have low or lower relevance from the page.

PageRank sculpting (SEO) and information architecture goals must both be pursued simultaneously for optimal results. Why?

In my opinion, the more PageRank points that Page 0 passes to the remaining pages that it links to:

• The more accessible those pages will be to search engine spiders
• The higher the PageRank will be for those pages, thus contributing to increased rankings (all things equal)
• The less PageRank points that will be “misspent” on low potential pages

The more useful Page 0 is to the subset of searchers who visit it:

• The more likely it will be to accumulate links in the future and the more PageRank points it will have to disseminate to the pages it links to
• It’s bounce rate will be reduced, thus improving the likelihood it will achieve or maintain higher rankings for the terms it targets
• The page will feel “smaller” and more intuitive to searchers who find it, which in our experience leads to higher site-wide conversion rates assuming most other pages on the website follow this same philosophy

Also note the role of prominence—another example of how something that affects Google search results also affects searcher relevance. Just as searchers do, search engine spiders give more attention to pages closer to the top (those that are more prominent). That’s because spiders won’t follow every link from page to page every time they traverse a site. Pages higher in a site’s hierarchy receive more spider visits and so have a higher probability of being re-indexed—and therefore of receiving more inherent strength in search engines.

Link allocation criteria

Ideally we want only pages with both high SEO potential and high searcher relevance (i.e., the “sweet spot”) linked from pages like Page 0 that have lots of PageRank to share—as illustrated below:

Even though they might not fall into the sweet spot, the most relevant links should usually be maintained even if they offer low SEO opportunity. In the past, low SEO opportunity links may have been NoFollowed as part of PageRank sculpting activities. Even though Google says there’s no benefit to NoFollowing these low opportunity links, if we believe their presence will improve the searcher’s experience, they could become a net SEO benefit. You do want to be stingy about linking to low opportunity pages, so consider using your analytics programs to weed out the truly relevant pages from the not-so-relevant pages and preserve your PageRank for high opportunity pages.

When we’re done prioritizing links that offer high SEO potential and high relevance to searchers, Page 0 might look like this:

PageRank points would then be allocated this way:

Now, high SEO potential pages have more than three times the PageRank points coming from Page 0 and there is less clutter for searchers to sort through once they reach the site. Techniques To Conserve PageRank Points

Here are some techniques to reduce the number of pages that a page links to—thereby saving PageRank points for key pages—techniques that also improve usability and conversion rates. These particular techniques resulted from conversion optimization testing we performed for particular clients, and may not work for you.

Reorganize the links that fall under a certain page on your site’s hierarchy. For example, you may find ten low SEO potential pages with very similar content. Rather than link to each one of these pages on Page 0, you may want to link to only one of them. Then link that page to the other nine.

Remove low SEO potential pages that offer little value to searchers you want coming to your site. You can then redirect them in an SEO-friendly way to more valuable pages. On category or product pages (or any page that appears deep in your site) remove the standard site-wide navigation. Replace it with several highly relevant links and content specific to that page. Provide a “continue shopping” or “continue browsing” button in the navigation for searchers who want to access other content on your site. Caveat: Test this tactic, as it might have an impact on usability; more on that below.

Perform usability testing and track conversion rates after you make your changes so that you know whether your changes actually have the effect you intended. Bear in mind that removing site-wide navigation from certain pages goes against standard usability practices. However, when done selectively, we have found it to be an effective practice when applied to some websites for improving both page rank and usability.

It’s also a good idea to perform usability testing and track conversion rates after you make your changes so that you know whether your changes actually have the effect you intended. Bear in mind also that removing site-wide navigation from certain pages goes against standard usability practices. However, when done selectively, we have found it to be an effective practice when applied to some websites for improving both page rank and usability.

A broader context

In my opinion, PageRank sculpting has a lot of value, especially when viewed in a broader context than just inserting NoFollowed tags. More than ever, your traffic volume depends on how well you prioritize high potential search phrases and pages. Both traffic volume and conversion rates depend on what searchers do once they actually reach a page. Do PageRank sculpting right and you’ll simultaneously achieve a stronger SEO result and convert more shoppers into customers.

When questioned, the panel agreed that links were important (since Google’s algorithm prioritizes their analysis of links so highly), but the general implication was that if you create great content the links will come.

It would be fair to say that sometimes good content gets linked to, but it’s rare. According to Google, as of late 2008 they’d indexed one trillion pages. So the probability that one new page of content would be found by a surfer randomly clicking from one link to the next on the internet would be one in a trillion. Man, those are not good odds.

There are many ways to help good content get linked to, but the practical reality is that these same techniques are more often used to help all content get linked to, regardless of the content’s quality. So SEO is just as much about creating “great” content as it is about being great at getting content linked to. I am not suggesting that the quality of the content doesn’t matter, but that the purist approach of simply putting great content into the world and waiting for unnamed altruists to link to it on merit alone relies too much on luck.

Five steps for creating & distributing content

When trying to improve organic SEO, creating great content is just as important as doing a great job of distributing it. These tactics, however, should not be pursued in a vacuum. Instead, you should approach SEO using a holistic strategy. Here is a sample five-step plan that you can customize to fit your company’s objectives.

Step 1: Search Google Trends for keywords that relate to your business that have received heightened attention recently (either because of seasonality or other special events). On a search for “outdoor chairs,” for example, you will see that the number of people searching on this keyword spikes every summer. If you sell outdoor chairs, now would be a good time to target that phrase aggressively.

Once you’ve chosen a topic area you are ready to move on to Step 2.

Step 2: Use the Google keyword tool to find terms related to the concept of “outdoor chairs” to write about. Through appropriate analysis you may find that the terms outdoor chairs, lounge chairs, wood chairs, and folding chairs are relevant to the products you sell, and that they get a lot of search volume.

Refine your target keyword list and move on to Step 3.

Step 3: Search Twitter for the keywords you selected in Step 2 to see what your target customers are talking about in order to get content ideas. @Macktacular is talking about painting children’s Adirondack chairs. @endlessrant seems to need outdoor chairs in a hurry. @jherskowitz is looking for comfy chairs to fit a rooftop bar. Try to pick content ideas that correspond to the keywords you decided to target in Step 2 because you know there is substantial search demand for those keywords.

Once you have your topic, target keywords and specific ideas you are ready for Step 4.

Step 4: Create your content and be sure to include the keywords you identified in Step 2, while also being conscious of what you’ve learned from browsing Twitter in Step 3. Popular content and articles are highlighted on Digg.com, so you may spend some time there browsing popular content that relates to your topics to learn how to structure your article’s title and body copy.

Your article is now written and you’re ready to focus on distribution.

Step 5: Determine a concrete “distribution” goal for each piece of content you’ve created, whether it’s an article, a how-to guide, etc.

Here’s an example:

Now it’s time to get to work. You are ready to distribute your content in accordance with your plan and enjoy the benefits of creating great content and leveraging it to increase traffic, search engine rankings, sales and brand awareness.