Asking the right questions
Lean Evangelist, UX Expert, Master of Experiments
Give your customers permission to be negative so that they can be honest.
Bring three to five questions and then use the journalistic pyramid.
Any question that can be answered with a yes or a no is probably not useful.
Lesson: Finding Customers with Cindy Alvarez
Step #5 Questions: Asking the right questions
There's absolutely such a thing as too many questions. I, like everyone else out there, when I first started doing this I would stress over a list of like, "Is this right? It's 20 questions, 30 questions, are these the right questions? Which order should they be in?" And I spent all this time stressing over it and now I just like to tell everyone, "You don't need to do this at all." You don't need 20 questions. In fact there's very little point in having them and having them can actually run counter to you because seeing that list, you will have a tendency to hurry people along so you can get to the questions.
Sometimes the best customer development interview is one where you never got passed the "Tell me about" because that's great. That means that you get gold on that first question. There's no way you'd want to stop someone having a productive conversation about a problem that they have that they're getting really excited about so that you can ask some other question that might be not very relevant at all. So I recommend people go in with five questions of which one is "tell me about" and the other is "If you had a magic wand and you can kind if pick three other things that make sense." It doesn't matter that much what the questions are as long as people are talking about their experiences and problems.
Once people start talking, you're going to spend more time asking for detail and kind of using the journalistic pyramid. How often does that happen, when does that happen, who does this involve, why do you think this is a problem, and that's going to talk most of your time.
The right scope for your interview depends on where you are as a company or as a product. So in the beginning you're really trying to figure out is there anything here? Is there anything worth continuing for which is basically is there a problem that you want to solve, yes or no. And you're trying to get some detail on that but if all you've got out of your customer development interviews was, "I talked to 10 people and they were all really empathic that they have this problem." That's good, you've succeeded, that was enough. You technically don't have to ask anything else. As you go on, you're trying to hone that in.
So in the beginning it's like very binary. Is there a problem? Yes or no. It's like once you go down the yes path then you start kind of treeing off which is, "Okay, given that there is a problem, what potential solutions make the most sense? Where is they're the biggest problem? Where are the biggest friction points? Where are the things where people would see the most value?" Because in any problem there's lots of solutions, so you're kind of starting to figure out which area of solution should we start going down.
And so then at that point your scope might be, "Well, it seems like where we could save people the most problem and create the most value would be improving this part of the funnel. So we'll start concentrating there." And so then your scope might get even tighter, it's like, "Oh, this feature now you're maybe thinking about how much people pay for it?" Which you could never ask directly because no one can tell you, but you start thinking about like how much do people value this or in what time scope do they need to fix this problem or how would they feel confident that it was fixed, like at what point would they go, "Yes. I feel good about this." And so then you're getting into a tighter scope and of course as you get features you're going smaller and smaller and smaller.
Probably the most common mistake is the leading question and the only way I put it is the yes-no question because no one ever thinks they're asking leading questions. That was neutral, right? Any question that can be answered with a yes or a no is probably not useful because people will either answer it politely or they just don't know. Would you use this? Yes, no, you don't know. Would you buy this? Even things like, do you understand this?
One of the things that I often say is that having built software in 15 years, people's understanding of something has very little basis on whether they'll use it or not. Most people who use all the products I've ever worked on have no idea how they work and yet somehow they're using them everyday. So not a relevant question. Do you like this better? Just none of those were good. So any yes-no question can be re-framed somehow and do a non-yes-no question and that's a very difficult skill, but you can do it.
Making it okay to say negative things is one of the big challenges. Especially I think in the United States we tend to be fairly forthright. I definitely heard from some European customers in start-ups that people are more polite, Japanese especially too. People are very polite. They do not want to say that things are bad or express problems with their lives. So a lot of times what we use is the other people have told us which is just to give permission and a lot of times I'll recommend doing this towards the beginning of an interview. It's just to say, "Other people have told us they have a challenge with this. Other people have told us they are very frustrated..." especially with existing customers. Other customers have told us that this part of our product is really terrible."
And once that permission has been granted, people are much more likely to kind of jump on board and then as people give criticisms, it's just like you would do with a child or a pet you kind of encourage them like, "Oh, it's so good to hear that." And even kind of saying sometimes it's really challenging for people to be honest with us and we can't get better unless we know what we're doing wrong. Being very explicit about that is really helpful because human psychology is working against us in a lot of customer development and so we have to kind of go ahead and tell people. People are more likely to be positive but to honest that's not that helpful. So anything you can tell us that's negative will be more helpful and then people want to please so then they can actually be forthright and say, "You know what, I hate this. This is terrible. I'm no good at this."
I have phrased a lot of times that my favorite question is, "If you have a magic wand how would you change this to make it better?" And that is my way of overcoming people's tendencies towards incrementalism. So when you say that it's kind of silly. People often laugh when you say it. If you have a magic wand, like forget what's actually possible, then that frees them up to say things that seem ridiculous but then a lot of times aren't.
So you can imagine in the Uber scenarios, it was like, "I just wish I could call a taxi and know where the taxi driver was." That sounds crazy. And of course if you have an understanding of apps and GPS and you realize actually that's not impossible at all. But your average consumer might not even throw it out because it seems ridiculous. If you said, "What kind of solution would you like for X?" And you said something, "I would like magic fairies to come in and make this better." It just sounds silly. Most people won't say that. They don't want to sound stupid. And so if you go ahead and give them permission to say something stupid, then they will. And a lot of times that impossible thing that they mentioned is actually there's a kernel of truth to it and that's what you can go after.