The HourSchool Blog

How can companies ask better questions?

One core pillar of human-centered design is the idea of seeing the world through other people’s eyes, gaining empathy for their worldview, and meeting them where they are. Many companies are getting on board with the idea of putting the user or customer at the center of their business and design decisions. One good place to start is actually talking to people, and design research uses qualitative methods to gain insights from interviews, observations, and workshops.

One useful thing I’ve found in teaching or training others in design research methods is that I’m forced to externalize and synthesize my own habits. Because they’re ingrained in the way I think, I take for granted certain important aspects of the work. One important thing that needs to be learned and continually practiced is asking good questions. You can use all the fancy design research techniques you want, but they’re only as good as the questions you’re asking and the mindset you’re asking them with.

I was a journalist throughout high school and college. I worked on both print publications and with broadcast media shooting film for tv and recording audio for radio. I learned a lot of interview techniques and interviewed a lot of people. I even took one course on conducting oral histories. We worked on writing good questions specifically and learned from firsthand experience when we held bad interviews that led to uninspiring stories filled with boring quotes. Although a journalist hunting down a good story and a design researcher seeking human insights have very different objectives, my time as a journalist definitely trained me to be critical of the questions I ask and why I’m asking them.

QUANTITATIVE & QUALITATIVE RESEARCH

There are two basic buckets of research: quantitative (hard data driven and deductive) and qualitative (inductive and considers multiple variables, their relationships, and context). An example of quantitative research is a questionnaire sent out to a large group of people. An example of qualitative research is conducting longer interviews with a small set of people. There are advantages and disadvantages to both, and a thorough research project may need both types of research methods for different types of data — always depending on the goal of the research efforts.

We (sociologists, anthropologists, design researchers, and others) choose methods of qualitative research to “[enable] the researcher to give the subject leeway to answer as he or she chooses, to attribute meanings to the experiences under discussion, and to interject topics. In this way, new hypotheses may be generated.” (1)

It’s key to remember that with generative design research, you want to stay open to new insights and new hypotheses that come from the participants themselves. When I am conducting design research (particularly generative research), I want the interviewee to tell me their story and guide me whereever they want to go. I cannot predict where their stories will lead, and I need to make sure I let them tell me what is important in their lives, in their contexts, and in their words. To do this, I start by asking a lot of open-ended, non-leading questions.

OPEN-ENDED & CLOSE-ENDED QUESTIONS

Close-ended questions prompt short, concrete answers like yes, no, or a number. Most surveys and questionnaires are comprised of close-ended questions. Open-ended questions leave the interviewees room to answer in the way they choose. Open-ended questions allow interviewees to direct the conversation. Again, you’ll need a mix of both open-ended and close-ended questions for a good interview (depending on your focus/goal), but err on the side of more open-ended ones. An easy way to create more open-ended questions is to be aware of the way your questions begin:

Closed

1. Did you…?
2. Have you…?
3. Are you…?
4. Do you agree/disagree…?
5. When did you…?
6. How long…?

Open

1. What…?
2. How…?
3. Why…?
4. Tell me about…

LEADING & NON-LEADING QUESTIONS

However, you can still ask all “why” questions and still be asking leading questions, which can be particularly limiting or even harmful during an interview. A leading question includes a presupposition or indicates what the interviewer wants to hear in the answer. The interviewer ends up directing the conversation instead of the participant. For example, any question that includes a proffered reason is usually leading: “Did you choose that program because your friends were already members?” It’s really really easy to ask leading questions, because most research is designed to answer questions or test hypotheses about new or existing programs, services, or designs.

Consider the following question in the context of a new business trying to discern what is important to potential customers. Though they will try to implement “green” practices, it is not the main selling point or focus of their business model. But they want it to be a part of their research because they have a hypothesis that the “green” messaging will win them some customers. One of their questions is…

Would you consider a business that uses solar or wind power over one that doesn’t?

First, how can we change this close-ended yes/no question into an open-ended question?

How does a business’s environmental impact or sustainability efforts affect your decision as a consumer?

That still pre-supposes that the interviewee cares about sustainability in the first place. Even if a business’s “greenness” doesn’t normally affect their decision, they’ll probably still provide some kind of answer and might give something that will make them sound good. This will skew the research data. Thus, an even more open-ended question:

When was the last time you made a [similar business decision]? Tell me more about your thought process and how you came to a decision.

This degree of open-endedness allows the participant to steer the conversation. They may not even mention sustainability at all. Though you’ll need a variety of questions with different degrees of open-endedness in a single interview, always try to work backward to find the broadest, most open-ended question you can start off with. In a brainstorm of research questions, it’s useful to include the very specific questions that you know you want answered, but know you may never ask them specifically of the interviewee. They can help you guide the interview once you’re in the middle of it if the participant has taken you to topic areas you’re interested in by mentioning sustainability FIRST.

Of course, your question can also be TOO broad. We at HourSchool have learned that it’s really hard for people to answer, “What do you want to teach?” It’s hard for people to have an answer off the top of their heads if they haven’t been thinking in those contexts or in those terms. Luckily, there are a hundred other ways to get at the same question.

1. What are you passionate about?
2. What do you geek out about?
3. What have your friends asked you to show them how to do in the past?
4. Do you find yourself automatically helping people with something when they’re stuck? What is it?
5. What’s something you’re good at, but you want to learn more about yourself?
6. What do you do for a living?
7. What side projects do you have?
8. What are some things you’ve done in the past year that you’re really proud of?
9. What online and offline communities are you actively a part of?

Here’s some advice from Recording Oral History: A Practical Guide for Social Scientists that jives with my mindset:

“A workable strategy is to use a broad question at the beginning of a line of questioning and then pick up on what the narrator says by asking more specific questions. The advantage of the broad question coming at the beginning is that the narrator follows his or her own thought processes or paths of association. You can learn much that you did not even guess about before the interview, including a new framework in which to view this topic. Rob Rosenthal, who interviewed people in Seattle to find out how their experiences in general strike had changed their lives, explained the advantages of a broad question: ‘Letting people talk about their worlds with as little structure as possible is a good way to see things through their eyes, and ensure against interviewer bias.’”

WHY IT’S IMPORTANT

One core pillar of human-centered design is the idea of seeing the world through other people’s eyes, gaining empathy for their worldview, and meeting them where they are. The mindset of respecting and valuing other people’s experiences, mental models, and worldviews without imposing your own will help you gauge whether you are asking the right types of questions. A quick check of the types of questions a company asks of its users is a good gauge of where they are on the spectrum of human-centered design and whether they’re walking the walk.

How much leeway do their questions give to participants to shape the answers (and thus their products, programs, services)? Are they mostly sending out surveys and questionnaires asking close-ended questions? Or are they talking to people face-to-face and asking open-ended questions? Are they open to people’s stories and experiences and questions? Or are they looking for specific answers to validate their hypotheses? Are they open to being surprised in the field, and are they ready to start with those learnings and run with them?

How can organizations and companies ask better questions?

Whether they’re aware of it or not, the types of questions a company asks sets the stage for the type of work they do and sets the culture for the type of relationships they have with their community.

-Christina Tran