How to make 100% sure SE don't follow outbound links

Some business partners with questionable sites demand being linked...

[ltrotta]

Hi all!

I'm no SEO expert, as I'm "just" a coder, but a client (and friend) has a problem and asked my advice -- and in turn I'm asking yours.

He has a web site with good rankings for his niche keywords. His business is both on-line and "off-line", in the real world, so to speak. He has a couple of business partner that have web sites with areas that some people would find questionable. These partners demanded to be linked from my friend's web site, and he can't refuse as he needs to do business with them.

He asked me what's the best way to link to these sites without creating not even the slightest association in search engines, and without being penalized because of the "technical" way he linked to them trying to make the SE ignore the link.

He's afraid that using "nofollow", redirects, just showing the url without the href tag, jscript, robots.txt might not be enough, as SE have technology to follow a link and create an association.

What should I recommend?

Thanks a lot for your good advice!

[Receptional Andy]

Hi ltrotta,

These partners demanded to be linked from my friend's web site, and he can't refuse as he needs to do business with them

My suggestion would be that he tell them he doesn't want to link to their sites as he feels that may risk his own search engine performance. If the sites are indeed 'questionable' then I wouldn't see this as a deal-breaker. But I guess this depends on exactly what the business relationship is, and why a link seems to be so important to them.

Certainly, any technique designed purely to prevent search engines from interpreting a link carries with it risk, apart from putting the links on a page that search engines cannot access.

[edited by: Receptional_Andy at 3:53 pm (utc) on Aug. 29, 2007]

[ltrotta]

Dear Andy,

thank you very much for taking the time to write!

My friend is in the publishing industry, and those partners are niche "independent" publishers that have poetry as well as "adult" sections (books, not magazines -- we are not talking about pictures). He can't afford not to represent their poetry catalogue, which is his niche, and is not even concerned about the "real world" association with those books, but he's afraid that a search engine might easily equivocate some content... after all they can't make the difference between "adult" pictures from book covers, can they?

Also, he wouldn't be able to explain why he's not linking them from the "partners" page.

Question: do search engines read cookies? What if a link was displayed only to browsers that are marked by cookies? That would be an acceptable compromise...

[ltrotta]

Other possibility... what if the link was part of a "POST" form? Again,it doesn't have to "look good", it just has to be there in some form... sigh.

[Receptional Andy]

I'm going to stick to my guns unfortunately ;)

A link between two sites is an explicitly declared relationship between them, regardless of whether this is visible to a search engine when interpreting the page's HTML. Remember that they have access to other sources that can reveal even 'hidden' links (e.g. toolbar data, analytics data, human testing).

If you are absolutely determined to have a link that search engines cannot easily interpret on a page that is accessible to them, then anything using a technology that is difficult for search engines to understand will likely 'beat' their analysis of the page's HTML. You could use encoded javascript, for instance or flash. Or you could go all the way and use cloaking.

But I would definitely advise against this approach which to my mind is counter-productive. I may be somewhat dogmatic, but I think that if you aren't happy to declare the relationship, don't link.

[Receptional Andy]

According to the thread below, nofollow may be OK to use without any negative consequences:

[webmasterworld.com...]

nofollow'ed links are dropped out of our link graph; we don't even use such links for discovery

[edited by: Receptional_Andy at 3:07 pm (utc) on Aug. 30, 2007]

[ltrotta]

Thank you for pointing that out! That is, hopefully, the solution.

[martinibuster]

Or you can use JavaScript, probably better. With nofollow the search engines know about the link, they're just not counting it for ranking purposes. If you simply don't want the engines to know at all, then JavaScript will work better.

[Tastatura]

some possible approaches:
- can you cloak for bots (i.e. show the link when human visits the page, and don't show it when bot comes)
- use redirect and deny associated folder in your robots file

[JohnRoy]

No Follow can mean 2 things: either in the page meta tag (where all links are ignored, sort of)
or in the <a rel=nofollow> attrib.

There's a big difference between their results.
[webmasterworld.com...]

Agree with martinibuster.
Javascript would be the best solution.

[superclown2]

You could link through a .cgi script and hide the script from the spiders by robots.txt or even .htaccess if necessary. I do it all the time. I wouldn't trust a nofollow tag, not every spider follows it, as I understand it.