Double Negative Leads to Proof Positive!

Once we’ve formed an opinion, we tend to look for evidence that supports that opinion and minimize evidence that contradicts us. This is called “confirmation bias”, and it’s probably the root of most bad products, bad hires, and bad relationships.

So how do we avoid it? In any situation filled with subjective information, we have plenty of opportunities to “believe what we want to believe”. In customer development, we may find ourselves subtly altering the questions we ask or interpreting responses in the most favorable light. In interviewing potential hires, the same thing may happen.

One way around this is the Trick Question. Identify your bias, and then ask a leading question in the opposite direction. We know from social psychology that people are likely to agree with an interviewer out of politeness/deference to authority. So most people will only disagree with you if they feel strongly in the opposite direction.

Here’s a customer development example:

  • Suppose I’m really excited about an “Uber for babysitters” concept. Tap a button on your phone and a babysitter will be there within the hour!”
  • The bias: I’m likely to ask questions about times when a parent has needed care urgently, about instances where a parent had to cancel an event when care plans fell through, about the hassles of getting cash for the babysitter, etc. Most likely I’ll get a lot of “yes” answers, which may lead me to minimize the potential risks.
  • Finding the trick question: Identify the positive aspects and find a way to turn them into a negative. For example, speed of finding a babysitter means parents don’t have time to vet them.
  • Asking the trick question: “Tell me about the last time you used a new caregiver. What steps did you go through before you felt comfortable leaving them with your child?” or “Tell me about a time when you or someone you know pulled a child out of a care situation because it ‘didn’t feel right’?
  • Why this works: Both of the above questions assume fear and uncertainty around the “Uber for babysitters” model. They give permission for the parent to express skepticism, so you’ll probably get negative responses. But if your hunch was right, you’re likely to also identify potential solutions (i.e. “Jen babysits for my friend Dave’s kids, so I didn’t need to ‘vet’ her.” — > connecting social graph to your app may alleviate fears around caregiver trustworthiness).

Interviewing candidates has a different twist. An interview is short: it’s hard to tell if a candidate is just saying what you want to hear. (If you blog and tweet regularly, as I do, it’s particularly easy for someone to figure out what I’ll likely agree with.)

  • The bias: Introduce myself as director of user experience.
  • Finding the trick question: Look for an example of a product or service where customers experience short-term pain but longer-term gain (i.e. Craig’s List is ugly but it works if you need to find an apartment)
  • Asking the trick question: “Craig’s List is ugly. What elements would you prioritize in giving it an improved look and feel?”*
  • Why this works: My job title gives candidates permission to jump right into describing small visual tweaks or usability improvements — they assume I’ll care a lot about those elements (and I do). But a good product manager/designer/researcher should make sure they identify the problems before thinking about solutions — which means they should push back and ask why a redesign is needed, or what goals I’d hope to achieve.

Double negative leads to proof positive!

* FYI, I don’t ask this Craig’s List question any more, prospective Yammer employees.