|
Thursday, May 29, 2008
 As some of you know, I work with a startup that brings in a lot of user-generated reviews. UGC completion in a complex field is not terribly unlike the shopping cart process for a purchase or a form process for lead generation. There are multiple stages, you may have to log in or enter detailed info, and it takes more than a single screen or stage to get done. Accordingly, it makes a ton of sense for anyone involved in anything of this nature (stage-based user completion funnels) to optimize and test how to get users from A to B without any significant loss of detail or prequalification requirements.
Abandonment rates depend to some extent on usability, to some extent on user error, but also on the nature of the intent they bring to the table. Uncommitted users who may have stumbled into a site from a search are much more likely to get "fed up" and "stop" in the middle of the process, but beware of taking that behavior of an indictment of your process. What happens to users who are a little more committed?
Recently we ran a loyalty promotion that galvanized and incentivized Charter Members to participate more deeply in our site. Many of those users came on friends' recommendations, bound and determined to complete the process. Accordingly, our aggregate funnel abandonment rate plunged from the 20% range to below 10%, and in some segments of users, near zero. On those days, aggregate completions were also way up.
As the promotion wound down, "normal" behavior returned and the less-determined behavior of the "average" web user kicked in once more.
But we now know how to trigger apparently "superhuman" behavior. For an ordinary user, the completion event was a faint barbecue smell wafting towards them from several acres away. The hot dog cart rolls up the street, and whammo, they stop searching for that far-off grill.
For a superhuman user who has decided to complete the task no matter what, it's more like a mouse just inches away from a particularly pungent piece of cheese.
If nothing else, this proves that nothing is necessarily inherently wrong with your cart, form, or process, as much as you should focus on making them smoother.
Lesson: don't trust aggregate numbers on anything to tell you a true story. Think about different kinds of users, and you may discover something amazing.
Posted by
Andrew Goodman
View Posts by Category |
|
D'oh!
|
