Changing all my external links to rel="nofollow or redirect?

Major sites do it, so why not?

[bouncybunny]

I rarely go to Yahoo and check the "linkdomain:" inlinks to my main sites - I've usually better things to do.

However, after a while of not doing so, I have just had a shock. A huge percentage of links that were once ordinary static html inbound links, have now changed to either rel="nofollow, or work via some kind of 302 redirect mechanism. Some go even further and direct to a preliminary site exit page with a message along the lines of "You are now leaving our site...".

The worrying point is not that this may happen on blog links, or forums (legitimate use of the rel="nofollow tag IMO). But rather, this seems to have happened mainly with links from edited and 'controllable' content on huge 'authority' sites - major broadcasters, news agencies and so on.

It's hard to know how much of this is intentional on the part of the websites themselves (for SEO or anti-link rot purposes) and how much is simply the result of changing practices by the people who develop content management systems. But either way, it seems to me that, in many cases, the default practice now is to discount outbound links.

This asks a few questions, including how Google and others see this practice. Does a 5 year link to a site which changes to a 'nofollow' indicate that the site is no longer trusted? What about if dozens of these links change over the course of a couple of years?

This perhaps being the case, I am considering changing all outbound links on my sites to rel="nofollow. This is not simply a petulant act of 'retaliation', I'm bigger than that. But I'm wondering if new web 'standards' are emerging (perhaps by default, rather than practice?), and if it is simply common sense to follow suit? Indeed, I am worried that *not to do so* will render my sites non-standard?

[blend27]

But rather, this seems to have happened mainly with links from edited and 'controllable' content on huge 'authority' sites - major broadcasters, news agencies and so on.

I'd say this has more to do with them tracking the visitors paths of exit. Just my 2 Cents.

[soxos]

Many people seem to be changing to rel no follow these days - and I wonder how this will effect the algo, have google / yahoo shot themselves in the foot when they came up with this idea? After all doesnt there algo heavily rely on the measurement of links?..

If for example you have a few links from a big trusted news website, and then they become no follow, are they counted or not, should they be?

I find it hard to imagine if the links aren't already, that they wont be at some point in the near future - I think we will get to the point when the resuls will be more relevant, not less with their inclusion, which means they will have to allow page rank / trust rank to flow through these links or they will be effectively basing their results on a much smaller pool of data than they have available - I can not see why they would want to do that..

besides the no follow tag is old really (2004?) and the search engines have got much better of detecting spam since then, so do they even need it?.. But then again, no one likes back tracking, and perhaps that is the situation they are in right now.

[piatkow]

Many people seem to be changing to rel no follow these days

Many people seem to have picked up the idea that you could manage the distribution of "link juice" by using nofollows which is based on an out of date view of how Google works. There is no longer any PR advantage in nofollow.

[aristotle]

In Google's algorithm, a dofollow link from an obscure free directory counts for more than a nofollow link from the New York Times or Harvard University. This is exactly the opposite of what it should be. The whole thing is totally messed up.

[piatkow]

In Google's algorithm, a dofollow link from an obscure free directory counts for more than a nofollow link from the New York Times or Harvard University. This is exactly the opposite of what it should be. The whole thing is totally messed up.

Not at all, the bot is doing exactly what it has been told to do. The problem is the redundant use of nofollow by webmasters because of the mistaken belief that it somehow improves the PR of the site giving the outbound links.

[aristotle]

Not at all, the bot is doing exactly what it has been told to do

You misunderstood what I meant. I was referring to the algorithm that Google uses to calculate rankings. It doesn't give any value to nofollow links. Thus a nofollow link from the New York Times or Harvard University doesn't get counted, whereas a dofollow link from an obscure directory could be counted.

[FranticFish]

There appear to be indications that Google realise that they've shot themselves in the foot with nofollow, because they definitely are at least crawling nofollow links from certain sites (Twitter for one).

I could start a big rant about how much I hate nofollow, how it's destroyed the web's natural linking and completely failed to stop spam, and what a bunch of lazy idiots everyone who thought it was the bees knees was...

[bouncybunny]

because they definitely are at least crawling nofollow links from certain sites (Twitter for one).

Really? And are they assigning PR to those links also?

[FranticFish]

Don't know about the PR, but here's where I read about it - [webmasterworld.com...]

[jimbeetle]

Be sure to read P1R's explanation in that thread on how the Twitter nofollow links ultimately get followed:

I can tell you that NoFollow is just there for the first part of the equation. Once you Tweet, and the API bots pick it up, that NoFollow get's stripped and away she goes! I know, I've watched it over and over again. My logfiles are a clear sign of what happens immediately after Tweeting a link.

[BillyS]

The worrying point is not that this may happen on blog links, or forums (legitimate use of the rel="nofollow tag IMO). But rather, this seems to have happened mainly with links from edited and 'controllable' content on huge 'authority' sites - major broadcasters, news agencies and so on.

Do you have some examples of major broadcasters that are doing this? I just tried several sites and none I looked at are using nofollow...

Cool experiment potential! I'd say give it a try and let us how you make out. I'd really like to know if, or how far, your site drops in rankings.


_{Personally, I'd rather have someone else do this and tell me what happened. I'm thinking the site will take a nose dive...}

[bouncybunny]

The BBC doesn't use no follow (AFAIK), but in some parts of their vast array of sites, they use 302 redirects in links that used to be straightforward links. There are others.

You really think a site would drop in SERPs because it changed external links to no follow?

[FranticFish]

I wish they would, but personally I doubt it.

What would happen to your site(s) if everyone linking to you no-followed their links?

If you vouch for a link, then let the juice flow. If you don't vouch for a link, why have it at all?

[koan]

Sites linking to good resources without PR obstruction should be rewarded by search engines in terms of relevancy if we want to see the freely linked web of the past come back.

[BillyS]

You really think a site would drop in SERPs because it changed external links to no follow?

I do think there is a pretty good risk anytime you make a site wide change like this. Especially since it's not how Google intended the option to work.

Too many webmasters are worried about PR juice and try to hoard it. If a link helps your visitors, then leave it. If you're selling text links, then a nofollow might be appropriate.

[bouncybunny]

Especially since it's not how Google intended the option to work.

Well yes, but in practice it's how it is increasingly being used. As I said at the top, the concern is that it may soon be the case that providing ordinary links will be the exception.

I wonder whether any possible short term issues with sitewide changes will be worth it in long term?

[CainIV]

There appear to be indications that Google realise that they've shot themselves in the foot with nofollow, because they definitely are at least crawling nofollow links from certain sites (Twitter for one).

They always did crawl those links, the nofollow attribute only removes the destination link from the Google link graph when calculating value.

I would suggest you link out to websites that are authorities and great in your space / related space, and not be concerned about nofollow'ing those links.

Some websites nofollow links that are resources, but often because they literally have 50 000+ resource links across the website.

[BillyS]

I would suggest you link out to websites that are authorities and great in your space / related space, and not be concerned about nofollow'ing those links.

Sounds right to me.

[FranticFish]

Be sure to read P1R's explanation in that thread on how the Twitter nofollow links ultimately get followed

Whoops! Missed that on first reading. Sorry for the misinformation.

They always did crawl those links, the nofollow attribute only removes the destination link from the Google link graph when calculating value.

Matt Cutts in 2007: "for Google, nofollow'ed links are dropped out of our link graph; we don't even use such links for discovery"