Nofollow Test - Leaking Link Juice?

Many webmasters loathe to link out to other sites for fear of losing link juice and on the rare occasions that they do, many use the nofollow attribute on a link element to make sure they don't lose precious link juice.

My approach is to link out to relevant content where it is appropriate.

If your site is full of nofollow links you are effectively saying I can't vouch for anything I link to - how valuable do you think such a page is?

I have created a test page that links to two similar pages. One has nofollow attributes on all of the links, the other one doesn't.



Please link to the nofollow test page with the above code to help the test out.

I would expect to see the page without nofollow attributes rank higher but we will see what happens.

If Wikipedia is anything to go by, it doesn't appear to have a negative effect. It'll be interesting to see what results you get from your experiment, though it'll be difficult to determine that it is just the nofollow attribute responsible for any variation and not duplicate content issues, etc.

It would also be better if the experiment were repeated with a variety of domains/content/links, as well as having control pages in place.

Good points, thanks.

I was worried about the duplicate content issue but not sure how to get aorund it.

I could try it with several different pages and if there was a resounding difference it would be obvious. I would swap the links round to each different pages so that the first link alternated between nofollow and non nofollow page then this way it would be even.
If I did this enough times (lets say 10) then it should average out to 5 each or close to it if it was an obvious 0 - 10 that would be conclusive.

This is only a basic test but as the pages are new and are very similar I think the results (dup content aside) should give some results.

Anythoughts on how to improve or do another test would be good.

I am not sure your study design makes any sense to me.

You are trying to test if a website with no follow links going out of it will rank differently from a website with do follow links pointing out of it

So why in heaven's name are you trying to get us all to link to one webpage.

The point of a test is to have two things that you are running at the same time.

Also you have commented on links coming out of your websites, outbound links, but you seem to be trying to generate links into the webpage, inbound links!!!

Can you please break this down for those of us who just don't get your study model

I'm asking for links to the test page that links to the two different pages that link out of the site. They are linke dto very similarly so that same amount of "link juice" is passed to the pages providing a control factor.

If the page without nofollow leaks link juice then in theory it should rank lower than the one with the nofollow links.

In answer to your question

The reason is to get the test page indexed so the pages with the links can get found.


Links out of a wbe page don't get it indexed. Links to a web page do get it indexed, hence me asking for them.

so what is the outcome of your Nofollow link testing?

I believe someone has already done this experiment, I can't find that exact article but here are the details that I can recall,

1. They create a random site
2. Create a few no follow links to site using a specific non popular keywords
3. The site has no relation or any mention of those keywords used for the anchor test

End results, that site did rank for that term. So it does appear some link juice and weight is being passed even though there is a no follow attribute.

This test wasn't run by me so I cannot endorse whether this theory is true or false.

I only use nofollow when linking to the big sites after all most of them use this themselves. I have no problem dofollow linking to a site as long as it is reputable and not part of any bad community.

some say no follow is just a joke. . . I heard stories that they submit there site to no follow sites but google still count that link as backlinks. . .

I'm trying it now but still waiting for the update.