Omniture tracking code URLs creating duplicate content

[BrianCC]

My ecommerce company uses Omniture tracking codes for a variety of different tracking parameters, from promotional emails to third party comparison shopping engines. All of these tracking codes create URLs that look like www.domain.com/?s_cid=(tracking parameter), which are identical to the original page and these dynamic tracking pages are being indexed. The cached version is still the original page.

For now, the duplicate versions do not appear to be affecting rankings, but as we ramp up with holiday sales, promotions, adding more CSEs, etc, there will be more and more tracking URLs that could potentially hurt us.

What is the best solution for this problem?

If we use robots.txt to block the ?s_cid versions, it may affect our listings on CSEs, as the bots will try to crawl the link to find product info/pricing but will be denied. Is this correct?

Or, do CSEs generally use other methods for gathering and verifying product information?

So far the most comprehensive solution I can think of would be to add a rel=canonical tag to every unique static URL on our site, which should solve the duplicate content issues, but we have thousands of pages and this would take an eternity (unless someone knows a good way to do this automagically, I’m not a programmer so maybe there’s a way that I don’t know).

Any help/advice/suggestions will be appreciated. If you have any solutions, please explain why your solution would work to help me understand on a deeper level in case something like this comes up again in the future.

Thanks!

[BrianCC]

Thanks for the detailed response and confirmation about the canonical being the best solution. This definitely helps.

Some of the tracking URLs are actually being indexed. It doesn't seem to be negatively affecting anything right now, but I'd prefer to prevent any potential future problems if possible.

Thanks again.

[Dr-Pete]

I think the canonical probably your best bet here. You can solve it with a 301-redirect, too, but it's a lot trickier. If you're really running into trouble, parameter blocking in GWT is ok here. Again, it's not my first choice, but it's not a black-and-white issue (just ideal vs. not-so-ideal).

If your pages are truly static, you'd have to write a canonical tag for each one, but most sites at least have a shared header and some dynamic components. In other words, your 1000s of pages may only actually be a few physical pages of code. In that case, you may be able to add the canonical tags on as little as one template (with some code). Unfortunately, this is completely dependent on the platform you're on - there's no universal answer (and the code is completely dependent on your URL structure). You'll probably need some quality time with your coders on that one.

The first thing I'd do, though, is to monitor your site with the "site:" operator in Google, along with "inurl:s_cid". In some cases, Google doesn't crawl these tracking URLs (or knows they're common to an analytics package). If they aren't being indexed, you may not have a problem here.

[BrianCC]

Thanks for the response.

The article doesn't deal with my specific issue exactly, but it does suggest using a rel=canonical in similar cases (affiliate tracking).

Using GWT to block parameters is a useful suggestion too, but isn't "recommended as a first line of defense" according to that article. I'll definitely use it in addition to whatever is best though.

Right now, the canonical tag seems like the best solution. Does anyone have any ideas on implementing these across the site's unique pages dynamically using code? Is this even possible?

Thanks!

[DRSearchEngOpt]

I think a previous article deals with this pretty well.  I would read the whole article but also take a look at utilizing GWT to not index particular URL Parameters.  Here is the link and I hope it helps.

http://www.seomoz.org/blog/duplicate-content-in-a-post-panda-world