June 2009 Archives

Tracking TWITTER using GOOGLE ANALYTICS

At first I found it amazing how plenty of twitter applications are flooding over the web everyday, but looking at this phenomenon closer may be it's not so surprising. According to ComScore there has been an increase of Twitter traffic of up to 700 percent since last year and number of twitter users are growing everyday. Therefore it's not surprising to see so many twitter tools emerging everyday, and sometimes it can be difficult to find yourself in this tsunami of twitter applications.

I was mostly concerned about tools on tracking twitter traffic. For sure, there are a lot of 'cool' tools out there allowing us to track number of hits, geolocalisation and even do real-time traffic analysis but it was not always convenient to have data separated from my main web analytics software. My approach to web analytics is to have a consistent, reliable and integrated view of the traffic thus ensuring a clear vision on what is going on in my campaigns. Furthermore, after testing several twitter analytics applications I didn't find the statistics from the different tools meaningful enough. So I revert back tracking my tweets on Google Analytics (GA). It is much more powerful in my opinion when considering all the functions that we have on GA. It would be wiser to use them to track our twitter campaigns. For example, we can track and analyse goals and conversion rates knowing very well how conversion rate analysis is imperative today.

Google analytics by default shows traffic coming from twitter but wait.. it shows traffic coming from Twitter.com only and today most of the people using Twitter never even visit twitter.com! For that reason, I prefer to manually add some utm codes on my URLs and this allows me to track my tweets wherever they are, even if someone forward my tweets by email.

As you know, tweets that include a URL use some type of URL shortening service, like http://bit.ly or http://cli.gs, but we will be using the classic Tinyurl.com which shortens a URL by creating a redirect that is hosted on www.tinyurl.com. The trick is to add GA's campaign tracking parameters to our Tiny URL, thus encoding campaign info into the URL we use in our tweets:

?utm_campaign=blogpost&utm_source=twitter&utm_medium=micro-blog

Of course you can change the utm_campaign and utm_medium names to anything you like but be sure they makes sense since you will be using them in your analyses later.

Example:
http://www.itjoblog.co.uk/?utm_campaign=itjoblogpost&utm_source=twitter&utm_medium=micro-blog

After adding this code to the URL, we shorten it using TinyURL and finally use the 'shortened' URL in our tweets which will be traceable by GA.

Below is a picture on how the data appear in 'All Traffic Sources' report in GA:

This is very cool, but it's even nicer to use all the power of Google Analytics in your web analyses, for example you could use Advanced Segments feature to subdivide your visitors based on operating systems they used to reach your tweets (Windows, Mac, Iphone, etc). Or if you like you could create your own segment that makes sense to your campaign.

As said by Gail Ennis, senior vice-president of marketing at Omniture: "Online marketers recognise the importance of brand reputation management in the social media environment," and I believe that accurate twitter tracking is fundamental and can help a lot in social media campaign analysis.

8 June 2009 | Web Analytics | Joseph Volcy | 2 Comments