Aren’t numbers great? Personally, I’m a datahead from way back. Can’t get enough of ‘em.
And they’re especially suited to my world of the web and user research. Hey, you’ve got real-world interaction; tons of data points; no facilitator or test bias; instant, real-time information …
The one thing that’s missing, though, is the why. Why did 32% of all prospects drop off at that point in the application process? Why are only 48.5% of users ever getting to checkout? Why do only 2.36639% of site users ever go to page X? What are people doing when they’re spending that average 27.3544 seconds on page Y?
Now, that’s not to say that there won’t be plenty of ideas. It sometimes amazes me how open to conjecture, projection, and confirmation bias some numbers types (business folks, typically) can be in these situations.
I had a perfect example of this phenomenon just the other day. One of our web analytic types had put together an excellent presentation on my group’s online application process. He sliced and diced the data by date, time, user type, device, site behavior, application page … Heck, he might have even have had eye color and Zodiac sign in there somewhere. It was awesome.
At one point, though, he stopped and wondered about something. In particular, he was wondering why so many people would abandon the application at the BYB page. (This is the page, at the beginning of the application, where we tell users how long this will take, what they’ll need, and so on. "BYB" stands for "before you begin.") Not too surprisingly some business types jumped right in with some possibilities – all with different degrees of being half-baked, fully-baked, or not baked at all.
Luckily, I was listening in and could help everyone out. I often test our site’s product pages. Typically, I just have the user check out the product page and see if that product is “right for them.”
One interesting thing I’ve found over the years, though, is that users in this situation often click on the Apply button (which, of course, takes them to the BYB page). They usually comment that they just wanted to see what the application process was like, or that they were looking for some bit of information that they couldn’t find on the product page.
The analytics guy thought that was really cool. The business types, on the other hand, have never been anywhere near as excited (they’ve heard this from me more than once).
I think at least he and I recognized that this is never an either/or situation. Quantitative and qualitative help inform each other. Results should always be triangulated, however you might do that. You really shouldn’t have one without the other.
Along with Jim Lewis, Jeff is one of the great numbers guys in the field of user research
No comments:
Post a Comment