How to deal with inbound affiliate links with an intermediate file

[AlmostHuman]

Well, here: [support.google.com...]

Google explains how to deal with paid links, which is also supposed to work for affiliate ones.

Redirecting the links to an intermediate page that is blocked from search engines with a robots.txt file

While it's clear how to do it with links with an affiliate id parameter, we still don't know how to deal with affiliate links with a hash tag (#)

I know it's a common issue, but we still strive to find solutions to try to redirect them to an intermediate "nofollow" page which would then further redirect user/crawler to the target page.

Has anyone encountered this kind of problem? Could someone suggest a solution, no matter how sophisticated and convoluted it might be.

[aakk9999]

I presume the problem is losing the hash tag (#) when redirecting to intermediate page?

If you are not able to use intermediate page, what about nofollowing the affiliate link instead of sending it to the intermediate page? This is a second suggestion on the Google help page you cited above:

[support.google.com...]

Adding a rel="nofollow" attribute to the <a> tag

<ADDED>

The other way that crosses my mind is to have an affiliate link that uses your own query string parameter, which then gets turned to hash tag (#) in your redirect script.

So if the affiliate URL is:
affiliate-example.com/some-product?someparm=something#some-hash-tag
you link to:
/my-redirect-script?affiliateurl=affiliate-example.com/some-product?someparm=something&hashtag=some-hash-tag
and your redirect script then turns &hashtag=some-hash-tag into #some-hash-tag

Disclaimer: I have not tried the above.
</ADDED>

This reply may be a bit late and you might have solved your problem already. If you have, it would be great if you could share a solution in case someone else has the same issue.

[edited by: aakk9999 at 7:04 pm (utc) on Nov 3, 2013]

[JD_Toims]

I presume the problem is losing the hash tag (#) when redirecting to intermediate page?

I would guess it's more along the lines of the hash (#) not being available for a server-side language to redirect in the first place since browsers act on it, but do not send the hash (#) or anything following to the server when requesting the page, so any hash-based redirection would have to be done with something browser-side, like javascript.

The other way that crosses my mind is to have an affiliate link that uses your own query string parameter, which then gets turned to hash tag (#) in your redirect script.

It would almost have to be done the way you're describing, or very close to it.

[turbocharged]

Amazon uses rel=canonical to point to the appropriate non-affiliate URL and has turned their affiliates into link builders. Google appears okay with this, so maybe you should look into it.

[Sgt_Kickaxe]

Big brand sites receive visibility boost in Google

A few updates ago big brand sites received a boost in Google serps while mainly affiliate sites suffered a loss of Google traffic and stopped being able to outrank the big brands on product keywords as easily, remember that?

We also know that Google changed how they handle urls blocked by robots.txt, they are now indexed without a description when they used to be left out of results.

We also know that Matt Cutts floated the question of ignoring nofollow directives if Google feels the site being linked to is trustworthy. The technology is there.

Put it all together and what do you have?

You have Google ignoring nofollow tags to big brand sites that GOOGLE trusts while also treating intermediate links being blocked by robots.txt as if they are just a bridge. The link value jumps from the affiliate site to the big brand site despite nofollow AND robots.txt and it can do it through multiple jumps and even through multiple domains.

So how can you link to affiliate offers without guaranteeing that the aff program outranks you? Titles. Do not use the same product keywords in your page titles. If you are linking to green widgets your page had best not contain those keywords in the title OR in the link anchor text. This will not stop the value from passing, it will stop it from passing for the keywords you need traffic for.

Example: If you wish to help sell a few bottles of widget glue then don't write about widget glue, write about a related craft and within that article mention why this particular widget glue is the best to use for that craft. Provide a recommendation with non descript "you can pick some up here" type link to the widget glue page and interested people will hopefully buy the glue.

You will NOT get traffic for widget glue, not that you would anyway given how aggressive Google has become at showing products on the right side of all serps, but to the crafters who appreciate your guide they may very well try it out.

[Robert Charlton]

We also know that Google changed how they handle urls blocked by robots.txt, they are now indexed without a description when they used to be left out of results.

Sgt_Kickaxe - Your comments are not accurate. Google's treatment of references to pages blocked by robots.txt was discussed on WebmasterWorld as early as 2003, and it's been amazingly consistent over the years.

Here's a fairly thorough discussion from 2012 of the mechanisms and vocabulary involved, with links you can track back to the 2003 posts...

Pages are indexed even after blocking in robots.txt
http://www.webmasterworld.com/google/4490125.htm [webmasterworld.com]

[JD_Toims]

Amazon uses rel=canonical to point to the appropriate non-affiliate URL and has turned their affiliates into link builders. Google appears okay with this, so maybe you should look into it.

That could be done if the URLs requested contain a query_string, but some-page.ext#stuff is not a different URL than some-page.ext#other-stuff, so unless a search engine is explicitly told the page state changes based on the fragment requested via #! rather than # in the URL the canonicalization won't do anything.

[aakk9999]

You have Google ignoring nofollow tags to big brand sites that GOOGLE trusts while also treating intermediate links being blocked by robots.txt as if they are just a bridge. The link value jumps from the affiliate site to the big brand site despite nofollow AND robots.txt and it can do it through multiple jumps and even through multiple domains.

Whilst nobody is disputing there is a brand bias in Google, I think we should be careful when making a definite statements like the above without some solid proof.

It is worth remembering that in cases where intermediate pages that perform redirect are blocked by robots.txt, then Google would not know where the redirected link goes to and could not possibly count it towards the end-target (i.e. towards the site the affiliate is redirecting to).