We are so awash in hard data these days – just like we are awash in psychotropic drugs. Now, while both of them are actually pretty good at treating symptoms, we’re not always sure how they did so or how we got ourselves into a situation where something needed to be fixed in the first place. There’s definitely some why missing in both cases.
As an old-fashioned usability engineer, though, I place great value on the why. If nothing else, I figure that, if we know why something didn’t work, we have a much better chance of not doing that again on the next project.
And sometimes that why really needs to sink in too. I know, for example, with writers, it usually takes more than a few tests for them to finally realize that people really just don’t want to read their stuff. Once they “get it,” though, they can then incorporate some ways to make their copy more attractive – slimming it down; supporting scanning and skimming by using lists, bolding keywords, chunking, using more headers …
Would we ever have figured that out, however, just from combing through web analytics? I doubt it. Similarly, it’s nice that an SSRI is helping you sleep and feeling better, but it’s not going to be much good and getting at the issues you have with your mother.
That last bit brings up a very important point. Most therapists these days try to combine the two methods – prescribe a drug to handle the symptoms, then work on the issues so that the symptoms don’t come back when the drug is eventually taken away.
So it is in UX. Use your data to identify problems (a poor conversion rate, say). Then, use good, old-fashioned usability testing to see where those problems come from (e.g., too many steps). Finally, use data again to see if the solutions you came up with (fewer steps) actually addressed the problem (a better conversion rate).
Henry Mintzberg is a biz school prof at McGill
No comments:
Post a Comment