Computerworld published yesterday a comprehensive article about the discussion in the analytics community around the measurement of engagement.
If you read this blog on regular basis you already know what I think and if this is your first time here, you can read my quotes in the article itself.
I won’t get here to the full discussion, but I did wanted to talk about another metric that is mentioned in the article - attention.
During 2007 Nielsen and Comscore came up with their own engagement metric. For them engagement is the measurement of the time users spends on a web site or a page.
This is not engagement. This is attention. What’s the difference? There is a thin line there but I believe that attention is passive evolvement while engagement is active interactions (such as comment, posts, purchases, sharing, etc).
But put definitions aside, I also think that measuring time spent on a page, although important, can also be on of the more misleading metrics out there.
I’ll give you an example:
A few weeks ago I had to change the date of one of my flights with Delta. I went to Delta.com and it took me more than three minutes of staring on the page before I figured out how to change my flight.
If all we measured was the amount of time I spent on the page, we will probably conclude that this is a very engaging page/site with a great user experience.
In reality it was exactly the opposite. It was one of the worse user interface possible, a confusing one that just made me don’t really like the Delta brand.
Don’t get me wrong, it is still important to measure the attention metric (and we also do that at NuConomy), but you just need to be careful when you get to conclusions based on that metric alone.
