Natural Search Blog

PageRank Sculpting – The Nofollow Debate

At SMX Advanced in Seattle last month, Matt Cutts made his intentions clear when he advised SEOs not to waste too much time on internal page rank sculpting using the nofollow tag and instead concentrate on spending that time on creating useful content.

This threw the SEO world into a tizzy with lot of power SEOs blogging about their concerns. Danny Sullivan’s post on Google Loses Backward Compatibility was an interesting one.

Matt Cutts cleared the air in his post on PR sculpting on his blog. Rand Fishkin has posted a video on PR leak at SEOmoz. Bruce Clay has recorded Matt’s comments on PR sculpting on his post.

All this information notwithstanding, at this stage, it is not very prudent to read too much into the new meaning that the nofollow link has assumed with respect to pagerank. A year ago, if a page had 10 points of page rank, with 10 links (external or internal is not important) on the page out of which 5 are nofollowed, the 5 followed links got 2 points of page rank flowing through them.

But in the new scheme of things, in the above scenario, the 5 followed links get only 1 point each and the remaining 5 points of pagerank evaporate. The term evaporate as it stands today is very ambiguous and does not really explain many things.

My understanding is that there is not infinite pagerank to flow across the web. It is proportional to the number of pages in Google’s index. If we look at Pagerank as a pie, the sites with PR10, PR9 and PR8 constitute huge slices of the Pagerank pie. The remaining portion of this pie is distributed across the millions of other sites on the web.

If a PR3 site wants to increase its pagerank to the next level, then it has to create lot more pages of content and therefore own more virtual real estate than its competitor who is sitting on a PR4. Apart from this, other factors like the authority and trust of the domain, quality of backlinks etc also come into play. All these factors are clubbed together into the toolbar page rank of a site.

SEOmoz’s Mozbar tends to give a more accurate picture by breaking up the TBPR (toolbar pagerank) into its constituents namely pagerank at the page level, trust at the page level, page rank at the domain level, trust at the domain level and the number of extrernal links from unique domains.

Going by the theory that pagerank is not an infinite quantity, the law of conservation of energy in science states that energy disappearing in one form reappears in an exactly equivalent form. In the same vein, the page rank evaporating from the nofollowed links on a page must be compensated by distribution of the evaporated pagerank elsewhere on the page instead of disappearing into thin air.

If there were outbound links to good quality resources with the same topical focus as the content on the original page, it would help in the distribution of the evaporated pagerank and remain useful to the users. It also leads to better quality authoritative relevant content and reinforces Google’s vision – provide good quality relevant results to searchers.

Over the years, webmasters have been reluctant to link to good quality sources from their sites. The use of excessive internal pagerank sculpting to flow link juice to important pages on their sites using the nofollow tag has sidelined the “perceived less important pages” on their sites. The concept of pagerank revolves around linking. The new measures adopted by Google could be to encourage sites to link more to trusted authoritative sources in their domain of expertise.

The increasing incidence of use of nofollow tags in links to external sites has also contributed to this quandary. The original intention behind nofollow was to use this tag to link to untrusted sources. They could be user generated comments, paid links, signup pages, links to RSS feeds etc. But the sole purpose the nofollow tag has served has been to channel link juice to specific money pages on a site to make them more powerful so that they can rank well in the search results.

Google has also come across cases where inexperienced site owners have done more harm to their site by using nofollow injudiciously that they have blocked off good quality content from the Google index. This is to the detriment of Google’s vision in providing quality relevant content to users. This can be disputed as the good quality content can be linked to, from other sites on the web. Google still knows the existence of such pages and can even rank them on the weight of the anchor text of the inbound links.

Matt Cutts admitted that Google started evaporating pagerank over a year ago. There are lots of sites (especially large ones) who have done internal page rank sculpting on their sites to flow link juice to their most important pages. There has been no significant report of sites losing their pagerank in this period en masse.

There are two other important issues that hinge around the value of Pagerank. They are:

  1. The inclusion of pages in the Google index is dependent on a pagerank threshold value the page must accumulate to be worthy of being admitted into the index
  2. The crawling of pages on a site is also dependent on the page rank threshold

Assuming that the pagerank evaporates from a page with a good PR score due to presence of nofollowed links, would the subsequent decrease in the pagerank cause problems for that page to be crawled in future? Would excessive evaporation lead to the page thrown out of the Google index? It is very difficult to analyse such factors and your guess is as good as mine.

Danny Sullivan made a comment on Matt’s blog wherein he points out that all links are not created the same on a given site. Google uses its own techniques to measure the importance of links and their ability to flow juice. Webmasters in their earnestness to adopt pagerank sculpting use the nofollow on links which, if followed, might still get credit from Google’s algorithm. But their manual intervention snuffs it out.

In a subtle way, Google is maybe reminding webmasters and site owners that they should stop being search engineers by trying to channelise pagerank flow and concentrate more on their business, developing quality content and improving traffic and conversions.

Michael Martinez in a post on SEOmoz had an interesting comment to make. He said that the TBPR is not an accurate reliable measure of the pagerank of a site/page. Internal pagerank sculpting has been pursued by SEOs across the board based on the TBPR score. He says that a technique based on faulty measurement will only lead to inaccurate results. It is akin to chasing rainbows. This is offset by Rand who has clearly stated that internal page rank sculpting has worked wonders for his clients.

Matt Cutts weighs in when he says that it is highly desirable to plan a good site architecture right from the inception stage of a website. The architecture should take care of the site hierarchy by providing links to important pages using a natural linking style rather than use nofollow to achieve the same results.

With all these deliberations going on in the SEOsphere, it would be ideal for SEO practitioners to avoid excessive use of nofollow from now onwards in their future assignments. If they have used nofollow in the past on client sites for internal page rank sculpting and there has been no outcry to date, it is best to leave them as is rather than undo all the nofollowing. More importance should be given to the planning of the site/information architecture to link naturally to important pages on the site.

Again, it is early days to fully understand the effects of pagerank evaporation. Many SEO experts have been testing this and their pronouncements should be awaited eagerly.

Ravi Venkatesan is a senior search marketing consultant at Netconcepts, our Auckland pay per click (ppc) marketing company in New Zealand. He also posts regularly to the Online Marketer blog at

what is hcg

No comments for PageRank Sculpting – The Nofollow Debate »

No comments yet.

Leave a comment

* Do not use spammy names!

RSS feed for comments on this post. TrackBack URI

RSS Feeds