Dynamic URLs and SEO don't always meet well

Often when building websites, its easy to forget that someday they will run live, having to compete with many other sites for top spot. Even if all onpage SEO is taken care of what about the URL's the site generates?

Dyanamic content can cause real problems to search engines and users, especially if the URL's change frequently. I've seen this in many soluitions in different languages, but one example that Grails and Java developers may well have noticed is jsessionid's with session in the parameters in some cases (and this is not desired).

Not all URLs are equal

What we are discussing here is how the URL's created by your server application look, here are some examples going from best through to worst:

• superwebsite.com/products/acme-rockets (Best URL)
• superwebsite.com?category=20384 (Acceptable but not great URL)
• superwebsite.com?session=weiourpqwioerjkadfl;category=20384 (Worst Case - probably not indexable)

So we can see from the URL examples above, that the first URL is best, users can immediately see what it means and its very easy to index and never changes. Moving on to the second example, as a user we have no idea what 20384 conveys, but because the category never changes, it is indexable. However, the last example containing the session is probably not indexable and each visitor, in each session will get a unique URL. What are the consequences of this:

• Anyone who creates a link to this page will include the session ID. Therefore that link is now to a page that no longer exists, no page rank transfer is possible.
• Google records the pages it comes across, but these pages no longer exist when it returns, however another page with the same content does, so therefore it removes them from its cache.
• Users who bookmark the site on a social media bookmarking service share there session information with the world, this is because they have created a link that contained the session. Suppose that was an administrator with a session that remained logged in?
• You'll probably hit google filters such as the duplicate content filter.

When a site has parameters that change with every visit you may not get indexed at all, I've seen a site with PR3 in Google get absolutely no pages on that domain indexed; that's how serious an issue this can be!

For google there's a possible solution, but I've not tried it and would not like to rely on it, from Webmaster tools you can adjust parameters that are to be ignored under settings, parameter handling. 

In conclusion

So in conclusion, during development check that the URL's generated by your application are clean, make sure that there are the fewest number of parameters in the URL possible, and realistically on most top level category pages there should be none.

This article was written by SEO and Web services Croydon,  a full service company offering SEO services in Croydon, London, Surrey and Kent.

If you would like to reprint or translate this article, please contact me using the contact link at the top of the page.