Customer interviews: Don’t lead the witness.

So I have been doing a bunch of customer problem interviews.

But I was warned I might be leading the witness.

Turns out I might have been, and there’s a better way.

Here’s what happened:

When I chat to each potential customer, I go through the same spiel:

I explain that I am trying to discover if the problem I am working on really matters to people. Adding that if it’s not, I should probably turn my time and heart to something else. ;-)

I collect a little demographic info.

I paint the problem context.

Ask a bunch of open-ended questions about the problem.

List a few specific problem statements and ask the customer to score the pain from 1–10.

And wrap-up.

I thought was really good.

But…the feedback I got back from a couple of my colleagues (and my wife) was that I might be “leading the witness.”

“No I’m not!” I shouted (albeit in my internal voice).

In my outside voice I calmly asked:

“Oh :-( Am I? Tell me why you think that?”.

My buddies said the when I describe the problem context and gave four short but specific examples of the pain I see…that I was leading the witness and possibly biasing their answers.

I was really torn. I am doing short interviews 25-minutes and I thought if I wasn’t specific about the problem…we’d be on a wild goose chase: “Soooo tell me…how are you? What’s new?” I thought we’d never get to the problem.

But…once I thought I *might* be leading the witness I couldn’t stop thinking about it.

When in doubt, experiment.

So, this week, after sitting with one of my UX buddies, I stripped out my specific examples from where I paint the problem context.

With some trepidation, I toddled off to my next interview. I held my tongue and talked much more generally about the problem space.

I was wonderfully surprised and heartened that this change did two major things.

1. It left more time for the customer to speak.
2. When they mention the examples of pain (that I no longer mention), I know this is real and true and raw…and not biased or colored by what I said.

So…hat-tip to my colleagues (and my wife) who gave me feedback on the original interview content.

And…I guess what I learned is, the witness does not leading.

Cheers,

Simon

Happiness Nurturer