Monday, January 7, 2008

Knowing That There's a Problem Is Often the Biggest Problem

A reader I’ll call Mike writes to me thusly: “They made me the product manager at my company!! They are asking someone who has been a server side developer for over 25 years to design user interfaces. They won't allow me to visit customers or do focus groups. How in the !@#$% do they expect me to design UI and processes? I don't know how the customer actually uses our product (I don't think the previous guy did either) and I guess I never will. I think I need to re-read your book so I will know how to commit every error you talk about. Sheesh!”

Mike is in a fix, all right. But at least he knows that a problem exists. Smart guy that he is, I have no doubt he’ll make decent progress on it, once he reorients his brain to cope with the new problem set that he now has. Here’s what I wrote back to him:

“At least you know that there IS a problem with most UI’s, and you know that you don’t have any experience with it. Socrates would say that this knowledge makes you wise, and I would probably agree with him. Many project managers wouldn’t realize that their users weren’t like themselves.

“As to not allowing you to visit customers, though, I’d have a huge problem with that if I were you. How can you possible know if you don’t look? If it were me, I’d insist on the authority to do that as a condition of taking the job.

“Focus groups I am less hot on. They’re hard to do really well. Consider, for example, the problem of simply recruiting the participants. Unless you can simply order random people to participate, as, say, the armed forces could, you have to recruit and entice them. And the guys that you’ll get are the ones who like to participate in focus groups, and who like to talk about software, which skews it towards the technophilic side, often largely. It’s hard to get real information out of them, as they often want to agree with and be polite to their presenter. They might not want to admit that they couldn’t figure something out, or at least not quickly, so as not to appear dumb in public. The ultimate example of a focus group failure is Clippy. Focus groups loved him because they only saw him once or twice. He only got annoying after he popped up the first five or six times. Then they wanted to kill him and his creators. It’s hard to find out what the silent majority really thinks, as by definition, they don’t say much. Automatic instrumentation of programs so you can see what users really do is, I think the answer.

“I’d suggest reading The Inmates are Running the Asylum by Alan Cooper, as a first step. You would probably also like his About Face, 3rd Edition, which just came out. And probably the best book of first principles in user-centered design is The Design of Everyday Things by Donald Norman. All are enjoyable reads.”

I wish Mike the best, and I’m sure he’ll do a fine job.