Possible Duplicate:
Should <a> tags with href=“#” use rel=“noindex, nofollow”?

If I have a page


which has in it's HEAD

<link rel="canonical" href="http://www.example.com/blog/mypost/"/>

When I link to the non-canonical version of the page should I add rel='nofollow' that way Google isn't wasting time crawling pages it won't end up using. A further complicating factor is that, in this case the canonical case is likely only linked in the sitemap, since all other incoming links to the post page will have context.

Is this sort of optimization pre-mature optimization and thus a waste of effort. The thought came to me from: http://www.quicksprout.com/case-study-techcrunch/ technique #1.


1 Answer 1


As an SEO stand point in this day and age you should never use nofollow to attempt to shape Google the PR flow or to stop Google crawling those pages.

There is no robots.txt, meta or link attitude that stops Google from Crawling your pages, and theres many reasons for this. There is however ways of telling Google not to index and cache this pages.

It's also worth mentioning that Google does not obey by the nofollow when indexing and why the attribute sounds like it would - it doesn't. Nofollow will not stop Google crawling pages since many links on that page could be follow and in this day and age most outbound links on sites are nofollow. Nofollow should be used when you don't want to pass any of your rankings to their outbound site, and should never be used internally since it has been abused in recent years and Google has clamped down on sites that use it internally.

Google only flows juice between pages that actually make it into their index so with this you can actually tell Google what Pages you want to have the most Juice by not indexing those pages.

What you need to do is actually use a meta name on the page that you don't want juice flowing to and also you don't want Google indexing it.

 <meta name="robots" content="noindex, follow">

It is also advisable to use robots.txt blocking the same pages from being indexed.. But the meta tag works better than robots, while having both is the best method.

Including that within the header of the pages that you do not want to index will tell the search engines not to index this page, HOWEVER! the follow command you notice will mean that if someone links to that page even though its not indexed it will flow the juice to the rest of the site meaning if someone links to your page that isn't indexed on Google or any other search engine for that matter your site will benefit.

  • The pages I don't want in the index have rel='canonical' so the meta header seems redundant, does it not?
    – Owen Allen
    Jan 10, 2013 at 22:24
  • Canonical should be used on all pages to avoid duplicates ending up in the index, ie pages that are accessible via multiple urls, a example of this would be url.com/?=page2/ and url.com/page2/ both have the same content and one is the master. In this case you would use <link rel="canonical" href="url.com/page2" /> on 'BOTH' pages and yes this would remove one from the index rel="canonical" should be used to avoid duplicates. While noindex, follow should be for pages that you just dont want indexed, therefor no canonical is required. Jan 11, 2013 at 13:25

Not the answer you're looking for? Browse other questions tagged or ask your own question.