Standardizing Web Analytics
One of the sessions I was lucky enough to attend while in San Jose last week was Issues in Analytics. This probably comes as no surprise to you, but as a blogger and a member of Bruce Clay’s writing team, I don’t spend too much time straining my eyes looking at numbers or scouring through Web analytics data. And, frankly, this is a very good thing, because should Bruce ever require me to do so, I would most likely have to hurt myself. Numbers are evil.
Despite all the numbers, I have to admit that Thursday’s Web analytics session was super informative and I came away with a lot of lessons learned, one of the biggest being how hard is to calibrate the data search marketers are getting from their analytics products.
Search marketers are running many different analytic tools on their sites in order to compile the most complete picture as to what’s happening. They’re measuring all sorts of events to determine whether or not a certain offline promotion is working, where users are entering their sites, whether a particular product should have its own landing page to increase conversions, what pages are getting the most traffic, etc. This is all great until you dive in and realize that the four different analytics programs they’re using refer to the same thing in four different ways and produce four (sometimes drastically) different numbers.
This isn’t good for search marketers. It’s not good for the campaigns they’re running, for the success of their sites, and really, it’s not good for Web analytics as a whole. As the industry matures, things need to be standardized.
Based on that, I was glad to hear speaker Avinash Kaushik mention during Thursday’s talk that the Web Analytics Association had just released 26 new standard definitions for key terms in an attempt to promote consistency among search marketers, agencies and their analytics teams. Anything that helps search marketers and those unlucky enough to have to deal with search marketers understand one another and speak a common language is a very good thing.
The newly defined terms were broken down into four categories and include:
- Building Block Terms: Page, Page Views, Visits, Unique Visitors, New Visitor, Repeat Visitor, Repeat Visitor & Returning Visitor
- Visit Characterization: Entry Page, Landing Page, Exit Page, Visit Duration, Referrer, Internal Referrer, External Referrer, Search Referrer, Visit Referrer, Original Referrer, Click-through, Click-through Rate/Ratio, Page Views per Visit
- Content Characterization: Page Exit Ratio, Single-Page Visits, Single Page View Visits (Bounces), Bounce Rate
- Conversion Metrics: Event, Conversion
Hopefully now that SEOs, marketers and their agencies are working under a standard set of definitions, the various analytics tools will begin to accept these terms as standard, as well. There’s a real problem when Google Analytics refers to something one way and then for OmniTure or WebTrends call it something else. I don’t care what term they use, but everyone needs to get on the same page. How else are search marketers ever supposed to establish a baseline to work off? What happens when search marketers switch analytics products? They’re supposed to change their vocabulary along with it? No wonder search marketers and marketing agencies can’t understand one another.
Jason Burby, co-chair of the Web Analytics Association’s Standards Committee recently wrote an article over at Clickz about The Importance of Web Analytics Standards and said something that really stuck with me:
Standards help define industry benchmarks and allow people to more easily compare themselves to others, to establish best practices, and to improve site performance.
I couldn’t agree more. You can view the WAA’s new set of definitions (PDF) on their site.