Hi

I googled this and could not find a suitable answer.

Basically a site i have started working with has affiliate tracking variables are appearing in googles organic results.

So for example several of my pages are indexed by google as

http://www.mysite/index.html?lang=en-uk&target=x.html&utm_source=clixgalore&utm_medium=banner&utm_campaign=clixgalore&mid=30710

This is not great for two reasons.

1. My Google analytics results are going to be wrong/overstated

2. I am likely to be overpaying Clixgalore as traffic from these organic google clicks will be misreported as Affiliate traffic.. when its clearly not.

So three questions.

1. How did this happen? I presume Google indexed my affiliates. Would the "rel=nofollow" be the most appropriate fix? If not.. then?

2. wishful thinking but How can i get my URL's relisted correctly... i presume there is no way to remove all my links rapidly with these paramaters in them?

3. Critically. How do i prevent Google tracking tags (or any other parameter tags) being indexed again. wrongly.  

thanks all for any thoughts.

tricky one

Hi,

I doupt that the rel="nofollow" would be an effective way to restrict google. You may want to use Disallow in the robots.txt file based on query string, see example at http://sanzon.wordpress.com/2008/04/29/advanced-usage-of-robotstxt-w-querystrings/ . If I were you I would use the following snippet of php code to redirect googlebase to the correct URL if the utm_campaign GET variable is set.

$landing='http://'.$_SERVER['HTTP_HOST'].''.$_SERVER['REQUEST_URI'].'';
$search = '|(\.html)(.*)|i';
$replace = '${1}';
$landing= preg_replace( $search, $replace, $landing );
if (isset($_GET['utm_campaign']) && preg_match("/googlebot/i", $_SERVER['HTTP_USER_AGENT']))
{
Header( "HTTP/1.1 301 Moved Permanently" );
Header( "Location: $landing" );
}

I saw you used html as file extension. You may include php code to html files using php_value auto_prepend_file in the .htaccess file.

Regards

Zoli

Hi

I had exactly the same issue not long ago and I posted the question in the Google Webmaster forums. A couple of people suggested canonical URLs. 

I didn't try it as the problem seems to have sorted itself.  However, it sounded like it might do the trick by specifying a preferred domain for different URLs that refer to the same page. More info:

http://www.google.com/support/webmasters/bin/answer.py?answer=139066&ctx=sibling

or

http://www.google.com/support/webmasters/bin/answer.py?hl=en&answer=139394

Good luck.

I would use a combination of events - use rel canonical  on all pages - and on robots txt exclude indexability of those URL IDs - however this is only advisable if your normal pages are indexed higher than the tracked URLs - a while ago I noticed this issue with Orange and posted about it