In a hangout, one SEO professional asked John Mueller about nofollow, and whether they should be using rel=nofollow on internal links?
John answered that yes, they can do this. But, this is something where he thinks that, for the most part, it makes very little sense to use nofollow on internal links.
If this is something that you want to do, go for it.
In most cases, however, John explains that he will try to do something such as using the rel=canonical to point at your URLs that you do want to have indexed, or use robots.txt for things you don’t want to have crawled.
Try to figure out if this is more like a subtle thing that you have something that you prefer to have indexed?
And then, use rel=canonical for that. Or, is it something where you say “Actually, when Googlebot accesses these URLs, it causes problems for my server, it causes a large load, and it makes everything really slow.”
Perhaps it’s expensive, or whatever. And for these cases, John recommends just disallowing the crawling of these URLs.
Try to keep it on a basic level there. And with rel=canonical, obviously Google will first have to crawl that page to see the rel=canonical, but over time Google will focus on the canonical that the site owner has defined, and then Google will use that one primarily for crawling and indexing.
This happens at approximately the 08:05 mark in the video.
John Mueller Hangout Transcript
John (Question)
Okay, is it appropriate to use the nofollow attribute on internal links to avoid unnecessary crawl requests to URLs, which we don’t wish to be crawled or indexed?
John (Answer)
So obviously, you can do this. It’s something where I think for the most part, it makes very little sense to use nofollow on internal links. But if that’s something that you want to do, go for it. In most cases, I will try to do something like using the rel=canonical to point at URLs that you do want to have indexed, or using the robots.txt for things that you really don’t want to have crawled.
So try to figure out if it is more like a subtle thing that you have something that you prefer to have indexed, and then use rel=canonical for that. Or is it something where you say like, actually, when Googlebot accesses these URLs, it causes problems for my server, it causes a large load, it makes everything really slow. It’s expensive, or what have you. And for those cases, I will just disallow crawling of those URLs.
And try to keep it kind of on a basic level there. And with the rel canonical, obviously, we’ll first have to crawl that page to see the rel canonical, but over time, we will focus on the canonical that you’ve defined, and we’ll use that one primarily for crawling and indexing.
