How Is NoFollow Data Treated By The Search Engines?

In theory, the search engines don’t follow links with the NoFollow attribute attached. That’s what NoFollow means. However, anybody who has been checking backlinks for multiple websites (for example, if you have many SEO clients, prospective SEO clients, competitor websites, etc.) will notice that Yahoo lists many NoFollow links as backlinks (I wrote about this last year, too.). I have seen this at Google (I believe in Webmaster Tools, but my memory is not certain on this point – sorry). 

If the search engines index NoFollow links, it is possible they use the data (otherwise, why waste so much computing resources indexing them?), despite that purpose of the NoFollow attribute being that the links should not count in their algorithms. This post speculates on how the search engines might use this data. 

A Partial History of NoFollow

Seasoned SEO experts can skip this section. It is intended for newbies, and it is only partial because I am sure I am missing out some details.

Before there were blogs, there were guest books. Guest books were like prehistoric Web 2.0 . They allowed website owners to create some form of user interaction with otherwise pamphlet-like websites. They engaged the user. They created stickiness. Best of all, they were set-it-and-forget-it, so many website owners thought “why not?”

Spammers quickly learned that they could drop links in guest books, which were often unmonitored. The extent to which this was happening reached near epidemic proportions to the extent that serious SEO specialists were leery of leaving any links in guest books for fear of having their websites penalized for spamming. Search engines were concerned because any mass linking scheme threatens to skew the quality of the search results they present their clientele – the searchers.

The search engines were let off the hook by the website owners. Those who did not moderate their guest books were disgusted by the spam. Those who did moderate their guest books were frustrated by the spam. For a low- or no-maintenance tool, guest books were proving to be a pain without any obvious benefit (such as increase in sales).

In truth, blogs came along and offered a much better way to engage visitors in a two-way conversation. Blogs offered a venue for opinionated and chatty webmasters to engage with visitors, and the blog CMS was much easier to handle than an “articles” section on the website (especially because many bloggers found they could dispense with pesky technicalities like grammar and even staying on-topic.). Blogs also offered a much more obvious business benefit than guest books – search engine rankings, which could be translated into increased sales.

It wasn’t long before blog comment spam had replaced guest book spam. But this time, the search engines would not be let off the hook. Blogs had so many obvious benefits and so much more invested in them that, instead of petering out, they kept proliferating. Indeed, each blog spawns hundreds or even thousands of pages, each one fertile for dropping a spammy link in a comment. And many blog owners were (and still are) lazy, allowing comments to be automatically posted without moderation. NOTE: This blog is moderated, and I use a DoFollow plugin. If your comment is worthwhile, your link will count. If your comment is not worthwhile, sorry.

Many bloggers became alarmed at all the spammy links, and were worried that they might be penalized for linking to bad neighborhoods. That’s why the search engines created the NoFollow attribute. And if you believe that, I have some superb oceanfront property on the moon that might interest you for a surprisingly reasonable price.

In fact, search engines were once again concerned because as I said earlier any mass linking scheme threatens to skew the quality of the search results they present their clientele – the searchers – and mass automatedblog comment spam was showing no sign of slowing down.

The search engines gave everybody, not just bloggers, a simple means to indicate when an outbound link from their website should not be followed by the search engines because it is not a link in which they have placed trust. Basically, the whole point of NoFollow is to eliminate user-generated links from the algorithms, since those links cannot be considered as “votes” for the sites being linked to by the sites doing the linking.

So Why Are Search Engines Indexing NoFollow Links?

 This is a puzzle. If the search engines created NoFollow to tell their robots not to follow, obviously something has changed since then, because they are following. But do the links affect the rankings? Here are a few theories of how the search engines might be using the data. These are highly speculative, so feel free to throw in your own speculations into the comments below.

  1. One obvious theory is that the search engines are not using this data at all in their rankings.
  2. A second theory is that the search engines are using the links to determine relevancy (a link from a comment on an SEO blog to my website helps the search engines confirm that my website is about SEO), but that the links do not count toward link popularity or PageRank.
  3. A third theory is that the search engines have built into their algorithms a process for selecting which NoFollow links they should include in their algorithm calculations. For instance, they might choose to follow all Wikipedia NoFollow links, but no MySpace NoFollow links.
  4. A fourth theory is that the search engines use NoFollow external links to dampen their trust rating of a website. If a website owner has lots of external links that it is not willing to trust, that is one signal that the linking website itself is not all that trustworthy. Makes sense for MySpace. Bummer for Wikipedia (but I’ve voiced my opinion on Wikipedia’s abuse of the NoFollow attribute before).
  5. A fifth theory is that the search engines use NoFollow internal links to dampen their trust rating of a website. Unlike some of these theories, this one makes sense. After briefly experimenting with internal NoFollow internal links on one of my websites, I removed them all. Think what message it sends the search engines about the quality of your website if you say you can’t trust your own web pages.
  6. A sixth theory is that search engines do not use NoFollow links directly in their rankings, but that they are included somehow in a link profile establishing a website’s level of activity on the Web.

I would like to hear your comments and theories. I should note that I have not researched this post in any great deal, because it really is just speculation. I wrote it while my daughters danced last weekend, and there is no WiFi there. So feel free to add your theories and enlighten me and our readers if you know of any great sources that can shed some light on this.

You can easily tweet this post by clicking Retweet This


  1. Hello, I really enjoed your post, and i would like to share my opinion about some of your theories.

    It woud be interesting to understand more about nofollow links shown on google webmaster tool, becasue it actually very hardly show you normal dofollow links. This is the first time i hear of nofollow links in the tool and i'd like some more reference about it.

    About your theories, i would vote for the first one. This is because those links shall not even be shown, and even less then shall be taken into consideration by search engines.

    The second and sixth theories looks very resources-consuming without having a real porpouse to justify it. I seriously doubt that google has an internal ranking for relevancy and "level of activity on the web" that consider nofollow links. Expecially the second one, i've never heard of anything like that, and i dont thing it actually exists.

  2. The thrid theory is my second choice, because it actually make sense, knowing google and their constant blacklisting with no public explanation of criteria. The can pretty much see wiki's as content pages and penalize myspace and facebook's links for obvious reasons.

    I presonally disagree on your forth and fifth theory. I do not think that a nofollow attribute means a bad vote for a page. It's like not voting, it just tell hte crawler not to follow that link, it doesnt say the destination page is bad in any way. Nofollow links can be helpful to avoid duplicate links on your pages, like the one in your head nav bar and in your footer nav bar that are on same page. It's also useful for not spreading the pagerank of your homepage to your privacy policy and term and condition pages.

    By the way, nice post. Have a nice day.


  3. I'm not an expert SEO, but last year I read an article discussed about page ranking. As we both know on that time in order to gain the interest of search engines you can actually do some buying of links and put it to a web page you want to gain search engine trust. This prompted Google to favor "nofollow" links to avoid bare selling of links from a high PR site as passively share link juice and page importance to other pages which it links to. To avoid this, Google used the nofollow link relation.

  4. OK, just for argument's sake, why would you want to avoid duplicate links on your page. I know a reason – if you don't want the search engines to follow the first link (as it normally would) that says "home" or that is your logo, and instead follow the keyword rich link at the bottom of the page, which it would normally ignore (so we believe, but I am not sure this has been proven). Can you think of any other valid reason to keep the search engine robots from following that first link? Other than trying to manipulate what anchor text they follow? I just can't see the search engines considering that to be a valid reason.


    aha, you can chalk this one up to the search engines being lazy, lol, because, these are all posted and are part of the search engines.

  6. I am new to SEO. I still have a lot to learn. I found your articles very interesting. I am learning so much from them. I appreciate you taking the time to share your expert advice. The more I discover about SEO the more I like it. Of course, it is challenging, but I enjoy challenges. Your blog and website will definitely be added to my bookmarks.

  7. I didn't know Google was indexing NoFollow links. Do they carry as much juice as DoFollow? I enabled DoFollow and KeywordLuv specifically to reward relevant comments but I'm certainly no SEO expert and would like to learn more on this subject.

  8. David, I gained better knowledge, after reading this blogpost, especially why search engines indexing nofollow links…

  9. Great overview, I wonder what Matt Cutts would say about this?

  10. I'm a very ordinary blogger yet my opinion -after reading- is that maybe they use the combination of all the theories you mentioned to dofollow the nofollow links. I also noticed that Yahoo! (via SiteExplorer) dofollow links more than Google do; means that Google (Search Results) actually dofollow what's dofollowed and ignore what's nofollowed yet they might be shown in the (Webmaster Tools).

    Qes: I wonder do you think merging theories 1 & 2 make sense? Meaning they don't collect data but they check relevancy and PR.

  11. That also is possible. I think the bottom line is that best practicves are to go for DoFollow links, but don't totally ignore NoFollow links, either.

  12. I've heard that indexing of nofollow links has something to do with something called that "pipe effect". How it was explained to me is that search engines are not perfect and when their bot arrives at a page it starts running through the site very quickly. After all, it's got to get to the next site.

    In the rush, some of the nofollow links get counted as regular links without the modifier. Kinda like sipping water from a large water main pipe. This concept extends to duplicate content as well.

    You can publish an article to a bunch of article directories and supposedly, the search engines are suppose to take one copy of the article and rank it higher while all the rest are shoveled to the end of the results list if they display at all. But it doesn't always happen that way.

    So maybe we're just looking to deeply at an imperfect system on this one.

  13. According to what my common sense says, I think, Google webmaster tools shows all the backlinks that are actually present including the NO-follow link. These basically are the links where we can get any traffic from.
    They don't really use this data for SEO purposes. The data used for Google SEO mostly depends on the Do-Follow links. Yahoo appears to index and value NO-Follow links, but do people actually optimize their site for Yahoo? I really don't think so.

  14. I have found many external nofollow links inside my account on the Google webmaster tools. I guess they're using some special trick to validate the good comments and make them dofollow…

  15. I personally don't think they count for seo with Google but do for Yahoo. I'm sure you do not get penalised for them otherwise that wouls stop us all commenting on blogs and social media. Some very experienced seo swear that no follows do count towards your serps so I guess it will run an run.

  16. Hiy’all . First Of All wonderful page . I loved reading your post.Just wanted to tell you, I voted you up at digg . Enjoy

  17. Its said that no follow does not pass pr but suppose you comment on a blog and you used an anchor text SEO and get 100 no follow links for that keyword will the keyword benefit in the SERP. Or will it effect rankings in SERP and harm the rankings. I am not an expert so am really very confused. PLZ reply.