Let's talk

Call us: +44 (0)1603 878240

New Google Referral Strings

00:00 on Wed, 15 Apr 2009 | Analytics | 2 Comments

If you're not analysing your own traffic logs and using Google Analytics / Urchin, then you are excused; this update won't impact you. For those of you who are crunching your own log data, you'll need to be aware that Google will soon be changing their referral strings from organic search.

Today, over at the Official Analytics Blog, Brett Crosby announced that during this week, we'll slowly be seeing new Google referral strings being introduced.

Currently, if you perform a search on Google for "flowers" you would see a string like this:

http://www.google.co.uk/search?hl=en&q=flowers&btnG=Google+Search

 

After the update, a search for "flowers" would give you a search string more like this:

http://www.google.co.uk/url?sa=t&source=web&ct=res&cd=7&url=http%3A%2F%2F
www.example.com%2Fmypage.htm&ei=0SjdSa1N5O8M_qW8dQN&
rct=j&q=flowers&usg=AFQjCNHJXSUh7Vw7oubPaO3tZOzz-F-u_w&sig2=X8uCFh6IoPtnwmvGMULQfw

 

Nice, huh? The main thing you're going to have to worry about here is the change from "search?" to "url?", so make sure any analysis you are doing does not rely on picking out "search" - or it will break.

 

Why the change?

Google isn't normally one to make things more complicated than it needs to be. It's interesting that they're obviously going to track even more data from the SERPs.

There has been speculation that variables such as "cd" may be "click data", tracking the current organic ranking of the site when it was clicked. The "source" can show which vertical the result was discovered through (e.g News, Blogs, Video, Images) and there's obviously some encoded information on the end too.

Google obviously feels that this data is important enough to their search quality to make such large change.

 

What does it mean for SEO?

Aside from the implications if you're crunching your own numbers, this change should also sit in the back of the mind of the SEO community. Google has historically, weighed the vast majority of it's algorithm to off-page factors such as incoming link authority, relevance and velocity. These signals were held above on-page factors, because of their relative "independence" from the webmaster, making them harder to skew. If Google is going to enhance it's search quality by looking at user behaviour, it will need more data, potentially similar to the referral data we're seeing here.

Hopefully, it's always been at the forefront of your mind, but looking at your own (or client's) websites in terms of usability, quality content and added value is likely to have more relevance to good organic positioning than ever.

Comments & Discussion

(2 Comments)

Post a comment, your email will not be published, nor will it be harvested.

Jamr0ck

Jamr0ck • Years ago

Matt Cutts has now suggested on the Blogstorm post (http://www.blogstorm.co.uk/google-adds-ranking-data-to-referrer-string/#comment-136464) that if you experiment, it should be clear what the changes do - i.e. it passes ranking data in the string. He sounds pretty excited about the change!

Reply

Jamr0ck

Jamr0ck • Years ago

I think this will signify the beginning of the shift to AJAX results pages - Something that Google have been reportedly running smaller trials on for a while. It could be a considerable headache for those people that rely on automated rank checking tools, or scrapers - at this point, only time will really tell.

I've also read that the CD variables could contain some extra SERP data, such as search positions, which might indicate that Analytics (and Urchin, for that matter) will go on to integrate position reporting in to their products, potentially eliminating the need for screen-scraping rank checking tools, as well as the associated server load they generate.

Reply