WTF are rel=”dofollow” & rel=”noindex”

 

For some reason this keeps cropping up, and I thought I would write a small informative guide.

I know one major web 2.0 property that recently added rel=”dofollow” to links, and there is a massive site doing 40M+ visitors a month which last time I looked was using rel=”noindex”.

Rel=”dofollow” Does Not Exist

It is not a valid microformat, has never been proposed as a microformat and has absolutely no reason to ever be included within the markup of a page.

As far as I am concerned the origin of “Dofollow” is the name Denis gave his plugin to remove nofollow from WordPress comments back at the beginning of 2005.
No Nofollow has been around for just as long.

That means that dofollow is a product name that has become synonymous with the movement against using nofollow links for all user generated content.

I really should update my list of dofollow and nofollow plugins though I am still using Lucia’s Linky Love.

You should also be aware that using nofollow might not be a viable method of hoarding Google Juice / PageRank Sculpting.

The ultimate reason however for this not existing is it has no purpose – links by default are “dofollow”… Google and other search engines then decide whether they have value to their algorithms.

Not all links, even without a nofollow pass PageRank, or even help with indexing – one example of this is Google Buzz which I have had a test running on for months – as far as I am concerned links from Buzz do not pass PageRank, and can’t help with indexing.

Rel=”noindex” Does Not Exist

Noindex comes in various flavours.

Meta Noindex
X-Robots Noindex
Robots.txt Noindex

They don’t all behave exactly the same, in fact I would generally ignore Robots.txt noindex as it seems to be the same as disallow which does not keep pages out of Google’s index necessarily.

The origin of rel=”noindex”? Probably people confusing the use of “noindex, nofollow” as a meta directive, and thinking you can do the same at hyperlink level.

You Can’t

rel=”noindex” does not exist, and it is logical that it doesn’t exist.

If you have a document, there can be 100s of links to it. If one of those links has a link level directive on it, all the other 99 links would still count.
Nofollow makes sense at a link level – 99 votes instead of 100, but you can’t have a document indexed only 99% of the time.

Thus if you see a plugin which includes rel=”noindex” as one of it’s features, stay well away.

If you run a website receiving 40M+ visitors a month and are using rel=”noindex”… LOL

Seriously if you are using either/both of these, no harm done… Google/Yahoo/Bing ignore them, but it might be worth cleaning up your code.

 

Liked this post? Follow this blog to get more. Follow

Comments

  1. says

    Points to a frightening level of ignorance, that things like this can happen, particularly on such large, well-established sites, run by people that supposedly know better.

    Nobody need know all the “right” things to do, but one would wish that they’d at least learn all the “wrong” things… and then not do them, anyway.

  2. says

    User-agent: Googlebot
    Noindex: /

    will prevent Googlebot from crawling *and* Google’s query engine from listing your URIs indexed by 3rd party signals on their SERPs.

    No other search engine supports it, though.

    • says

      When I played around with robots.txt noindex in the past on short tests it didn’t seems to drop the pages as fast as a noindex on the page, (as in the page was still showing so I got impatient) whereas you would think it could be almost instant… just after the last crawl.

      Most of the time I have been doing big redirects on large sites there was a need for a 301 so neither could be used

      I have a few pages that I keep meaning to noindex so will do some cycling of robots noindex on them “now you see me, now you don’t” that might provide some data – so much easier to do a similar test with code on the page.

  3. says

    And then, as long as we’re having fun with myths. Let’s not forget that old favorite imaginary tag, the Revisit After meta tag. That one makes me stand up and start shouting.

    It just goes to show that html is pretty amazing. Why isn’t most of the internet broken what with us inventing our own code all the time?

    :-)

    • says

      I must admit I am less woried about things that really didn’t even have the “intent” of doing something.

      Thus is someone wants a page out of the index, rel=”noindex” isn’t going to do it, but the intent is there, thus their site isn’t structured the way they want it.

      On the big site I remembered using it I checked again last might, and they are using rel=”noindex” to link to some kinds of internal categorization.

      The only positive purpose would be if when looking at a page you wanted to know which pages point to other pages that use noindex – I could imagine that being useful in some very remote ways, plus then you might look to use a class instead.

  4. says

    Wait, those don’t exist?!?! ;)

    Yeah.. the shit gets quite annoying when people keep doing this. I wish Google had a damn flag they could put on there to show ignorance. ” – DUMBSHIT”

  5. Srinivasan says

    I was surprised to read the ‘dofollow’ and ‘noindex’ post. What a lot of trash the so called internet marketing gurus are teaching. My own misunderstanding on the subject is now clear.

    • says

      I don’t think this is anything related to “internet marketing gurus”

      I have seen some misworded statements on very reputable resources that I am sure if I actually mentioned them the author would be embarrased because they really do know certain things don’t exist.

      The problem is often that something gets picked up on forums as gospel, regurgitated on blogs and becomes some hideous myth that just won’t die.

  6. says

    I gave up on explaining SEO myths long time ago. You come to a forum, they start discussing about SEO, and it’s always the same thing. Some newb SEO comes in and gives “valuable advise” such as “You need to submit a sitemap to Google.” and “You are missing this keyword in meta keywords.”

    I just don’t have the energy to argue, link to Matt Cutts’ blog posts and videos and everything anymore :(

    • says

      On some things I agree, but when I saw cases with large sites I thought it worth a mention – it gives something for other people to link to as well which is always useful :)

  7. says

    The “noindex” attribute is being used by some Russian search engines, namely Yandex (http://www.yandex.ru/) which is the biggest and most popular SE in Russia. In particular it considers the tags (noindex)(/noindex) that state that the content between them should not be indexed. Thus, you can manually set what %% of document should be indexed.

    Didn’t ever find any example where those tags could be of any use though.

    • says

      Yandex is like this

      <noindex><a rel="nofollow" href="URL">Link Text</a></noindex>

      Yahoo have something similar with Robots-nocontent that was implemented as a class.

      I can think of one very good use, Tumblr blogs which are syndicating other people’s content heavily – the copied section could be ignored.

      From what I have read Google have no intention of adding something similar because uptake of the Yahoo version was very low.

      • says

        Very good remark! Copypasted content could be closed from the search engines, that makes sense.

        BTW, Yandex also takes into consideration the meta name=”robots” content=”noindex”.

  8. says

    I thought rel=”dofollow” was what spammers added to links dropped in blog comments (so that Google then ignored the subsequent nofollow rel added by the blogging software). Didn’t it even work for a while?

    • says

      I have seen that though I think it mainly fools toolbars not Google.

      The site I say it on switched from a nofollow to the rel=”dofollow” quite recently, which is why I spotted it.

      The big rel=”noindex” site has had it for well over a year.

      • says

        Yes, it definitely fools (s0me) toobars/custom style sheets. I thought I’d seen Matt Cutts say Google had taken some action not to be fooled. Can’t find it by searching so maybe I imagined that …

  9. says

    I’ve been getting pretty irritated over this nofollow tag in general lately. I understand the desire to eliminate comment spam or spreading page rank/value to the wrong pages, but there has to be some better away about this. It feels as if we are given this code to help alleviate a lot of the garbage from search results, yet all it’s done is create confusion.

    Anyway, I appreciate the informative post.

  10. iDealer Services says

    I am not a computer scientist, but there has to be a better way than using links as the primary determinate of which sites rank highest. The web has become one giant spam machine with everyone and their mother fighting for links.

  11. says

    My site got indexed but only 1 page, I have been now waiting for weeks to update my website keywords and other pages. It take s forever google to do anything to your site if it has no pr.0.