Sara Sodine Parr: How to Ask the Right Questions to Build the Right Thing

Sara Sodine Parr answers questions about asking questions. Image courtesy of Parr.

Sara Sodine Parr’s job is to ask the right questions. To find out what users think and need, and use that intel to inform how products are designed, built, and improved. If this sounds easy, it’s not. In fact, you’re probably asking the wrong questions as we speak. 

A Lead Researcher at Airbnb, Parr has seen firsthand how start-ups can benefit more than any other company by talking to their customers, especially as they try to figure out their product or product market fit. But those are also the companies that don't invest much time talking to their customers, or not as much as large enterprises.

Parr has launched a new course Build the Right Thing that digs into how you can get customers to reveal exactly what they need, so you can take the guesswork out of building great products. (The first session sold out, but you can join the waitlist for the second.) Here, she discusses how to ask the right questions, why humans tell you what you want to hear, and why the key to designing for the unexpected lies in support tickets. 

You note about how switching even a single word in a question can make a significant difference in how people respond. Can you give us an example? 

My favorite question is actually not a question. It's starting with the words, “Tell me.” Tell me about this behavior that you did. Or tell me more about this idea that you just shared. “Tell me” gets people to elaborate on what's going on. It’s the best way to not lead people, because the more that you say, the more you can impact or influence how they respond. 

My other big one is to avoid asking “Why?” That is the most obvious question to ask—and it’s what people default to—but “Why?” sets off a trigger for the person you're interviewing to make up some rationalization for what they're doing or feeling. Instead of asking “Why?” I try to ask “How?” or “When?” or “What?”. It elicits a more specific and truthful response.

What have you learned about human behavior from talking to people?

People are overly optimistic about what they're going to do in the future. People are also appeasing, so they say what they think you want to hear. Maybe you interview them to understand what their problems are, but they answer to try and make you happy and tell you they like your product or idea. 

That’s part of the reason why it's bad when people say that they want to talk to customers to validate their ideas. Validating your ideas is easy because people are optimistic and appease you by default. If you go into it with that goal, you're almost always going to come out with your idea validated. I try to focus on getting people to open up about their problems. 

How do you do that? 

I focus on past behaviors. Instead of asking, “What do you think about this today?”, ask them to take themselves back to the last time they did some action and walk you through what happened. That forces them to think about what happened in real life. You ground them in some sort of truth. 

If your product is live, you can also observe them using the product by giving them different tasks and watching how they use it. You will uncover problems they might not even be aware of. If you don’t have a product yet, another tactic is to watch them use a competitor’s products to identify problems in the competitive space. Then you can build something 10% better. 

Are there other questions we ask all the time that aren’t really effective? 

People want to understand from research how much someone is going to pay for something. Those are questions that you don't want to ask the customer because they really don't know. They're going to be overly optimistic about it, and they might say, “Yeah, I would buy it,” or “Yeah, I would pay that much money,” just to appease you.

What’s a product you’ve come across recently where you wondered if they did any customer research? 

I'm staying at a friend's place in New York, and they have an electric kettle that I was trying to turn on. There is a light on it that is round and it looked like a button, so I kept clicking the button. I was doing it forever, and it did not turn on. It turns out that there was a fancy-looking clear switch in the back that you had to push down to turn the kettle on. 

The lesson there is that if you make something look like a button, people will think it’s a button. If you’re overly inventive when you design a switch, people won’t know it’s a switch. Why? People use your product with intuition they’ve developed from every product they’ve used in the past. If you can, stick to the norms people know.

That reminds of a conversation where I learned that they have built bike lanes in parts of Europe at a different elevation than the road. They assumed cars would drift into the bike lane and designed to anticipate human error. 

Recently, my  husband was doing a road trip from Austin to New York. At one point, he left his key on the car roof, started the car, and drove off. The key flew off the roof and ended up in the street somewhere, but he didn’t realize that the key was missing until 100 miles later. I was like, How could the car not be designed to tell you when your car is 50 miles away, or even five miles, from your key?

What’s interesting about that is it’s hard to design for the worst-case scenario. You design for the likely scenario, and neither you nor your customers could have imagined the worst case. It’s important to review customer support tickets to look for the worst-case scenarios people have and solve them at scale. 

What’s a customer research best practice that you think has no correlation to how people operate in real life? 

My least favorite is customer profiles or buyer personas or whatever you want to call them. We should focus on people's problems and real customers. I don't like the idea of having fake profiles of customers who don't exist. There’s probably no single customer that fits all the weird descriptions we put in buyer personas.

Good personas do include information about real people's behaviors and attitudes, but, when it's packaged as a persona, people focus too much on the demographics. You could have a 50-year-old man and a 30-year-old woman use your product in the same way, so focusing on demographics is something that I almost never do. 

Outside of the research field, who is your favorite interviewer? 

Dax Shepard. He’s good at making people feel comfortable and that is an important skill for customer research. To get past all those barriers of people being overly optimistic or trying to appease, you have to make them feel comfortable. Dax gets vulnerable, which makes his guests get vulnerable. Having that friendly side is a good thing. 

If you’d like to read more from The Creative Factor—such as Morten Bonde’s story about reinventing himself as a LEGO Art Director while losing his sight or the profile of design trailblazer Sara Little Turnbull, who reimagined the bra cup as the N95 face mask—sign up for our newsletter.