Will Google count a page with a tracking parameter as separate page?

Or is it smart enough to know they are the same page?

[thrasher141]

I have a page called blabla.com/page.asp?id=1. I use Webtrends to track campaigns - ads and incoming links. They way that works is by tacking on a parameter to the URL, so my URL above would turn into blabla.com/page.asp?id=1&WT=linkID

That would be the URL I put on an affiliate site or incoming link. Will Googlebot see that URL, crawl it, and index two different pages? Or will it realize that both URLs are the same page?

I'm sure this has been discussed before but I couldn't find it. Thanks.

[trimmer80]

Hi,
Firstly I would never use the parameter "id", search engines are known to have had issues with this parameter name in the past.

Secondly google will see these as seperate pages. Google is smart enough to see these as duplicate pages and thus will not rank on of them but I imagine you will also lose the page rank benefits associated with the link.

[McMohan]

thrasher141, I have had some experience with this problem for a client that I do SEO for. They have their own web server and have some issues with installing a 3rd party web analytics tool. They track everything using these tracking URL, so a page within the site will read as www.example.com/company.htm?channel=example. The same page will read as www.example.com/company.htm?channel=portal if they have an ad in a portal. Plain ridiculous and not the best one to have for a client.

However, Google has managed to see right page in most of the cases. But on few instances Google has indexed URLs with tracking ID, and that I guess is due to the fact that such pages are linked from some portals with good PR. I have advised them to shift from URL tracking to a tagging based traffic analytics solution such as Google Analytics, which I guess is the only recourse. But then, I too would be interested if someone has a different solution.

[ALbino]

I was told a long time ago (and don't know if it's still true) that when using a referrer id (like for a reseller program) G will eliminate the dupe pages, but the incoming links won't really pass any PR to that specific page. For what that's worth.

[g1smd]

Make sure that your on-page scripting "tests" the URL of the page as it is being served and adds a <meta name="robots" content="noindex"> tag on all versions that include extra unwanted parameters.

Google will then list the page only once and only under one URL. Don't rely on them to choose which URL to list, you should prevent them from listing all of the alternatives, yourself.

PageRank will sort itself out.

The product page may be known as:
/widgets.php
/widgets.php?refid=1
/widgets.php?refid=2
/widgets.php?refid=3
/widgets.php?refid=4 (with these last 4 set as "noindex")
and each "page" will have its own PR. Assuming that every page has standard navigation linking back to your main index page, and to other main section indexes of your site, then all of those "pages" will pass at least some PR back to the rest of the site anyway.

[McMohan]

g1smd, I don't think I got you right. How can one add <meta name="robots" content="noindex"> to a page that doesn't exist? I mean the URLs with tracking code don't exist physically.

[g1smd]

Well, they must exist because when you go to that URL you get content and a HTTP status code of "200 OK".

Yes, it is the same physical page on your server that is returned for many different URLs that are used to request it. So, have a script ON the page that detects "by which name was this page asked for" and get that script to add the meta tag if the name includes any parameters, and omit the meta tag when it does not.

[kaled]

I too would be interested if someone has a different solution.

Ages ago, the use of url.html#trackid=refid was discussed. This negates the possibility of pages being indexed wrongly but it does raise a number of other issues.

Kaled.

[ALbino]

Do it Wiki style:

whatever.com/page:refid

AL.

[thrasher141]

Another solution I have discovered since posting this is to use some javascript on the link to add the referrer ID. For example <a href="http://www.site.com/page.asp?id=374" onmousedown="this.href += '&refID=blabla'">Cool Site</a>

Bots will see this and follow the href, but not do the mousedown part. Human surfers with javascript enabled will click this link with a mouse and be taken to [site.com...]

Obviously this isn't foolproof because some users don't have javascript and some don't use a mouse, but it will probably get 95% of hits.

[g1smd]

Is there a danger that Google will see that as being something in the same family as a "sneaky java redirect" becuase the onclick URL is different to the href URL?

[thrasher141]

Hmm, I hope not. I haven't heard of the "sneaky java redirects". I could see how Google might not like that, though perhaps their bot won't even process the javascript so it won't even know? This technique, using the onmousedown event, was recommended by a company that does web design, analytics, and SEO. So it seems like they would know their stuff. I'd certainly love to know if Google has a problem with it.

[kaled]

Google should not have a problem with this. There are legitimate reasons why you might direct js and non-js enabled browsers to different pages. HOWEVER, it has been said that Google follows full urls (http://....) irrespective of where the urls are located. That's something to bear in mind.

Kaled.

[g1smd]

Typo: java should have read javascript: [google.com...]