Why talk to people at all
Your own experience is only one sample. People who will use what you make have habits, frustrations and workarounds that you cannot guess from the outside. A short conversation can overturn an assumption that would have cost weeks of work.
User interviews do not need to be formal. A twenty-minute chat with a few people who match your intended users is enough to start. They will not prove anything about large groups, and a small number of conversations can mislead if the people are not typical. Treat what you hear as clues to check, not as final answers.
Who to talk to
Aim for people who actually have the problem, not just people who are easy to ask. Friends and relatives are convenient, but they may say what they think you want to hear. If possible, include a few people you do not know well. Try to include variety in age, experience and ability, since a narrow group gives a narrow picture. Ask politely and explain that you are learning, not selling.
Three to six conversations is a common starting point. Stop when you keep hearing the same things, and start again if the problem changes.
Prepare, lightly
Before the conversation, write down what you want to learn. Your problem statement helps. Then prepare a handful of open questions and a loose order, and treat them as a guide, not a script.
Choose a comfortable setting. Ideally, meet where the problem happens, such as the kitchen, the workshop or the garden, so the person can show you instead of just describing. Ask permission before taking notes or recording, and keep the information private. If you plan to use quotes in public, ask first.
Ask open questions
Open questions invite stories. Closed questions invite yes or no. Compare:
- Closed: “Do you find it hard to open jars?”
- Open: “Tell me about the last time you opened a jar.”
Useful starters include:
- “Can you walk me through how you do this now?”
- “What was the last time that happened?”
- “What is the most annoying part?”
- “What have you tried to make it easier?”
- “What happened next?”
People are generally more accurate about what they did than about what they would do. Asking about the past (“last time…”) tends to give richer, more reliable answers than asking about the future (“would you use…?”). Hypothetical questions about what someone would buy or use are especially unreliable.
Avoid leading questions
A leading question hints at the answer you want. “Wouldn’t it be great if the handle were bigger?” nearly guarantees agreement. “Don’t you hate it when…?” pushes people toward complaining. Try to keep your wording neutral and drop pet ideas from the conversation. If you describe your solution early, people tend to respond to it politely instead of talking about their real situation.
Other habits to watch:
- Do not finish sentences. Silence is useful. Wait a few seconds and people often add more.
- Do not argue or defend. If the person dislikes your concept, ask why.
- Do not pile on. One question at a time.
- Do not correct them. If they use the product “wrong,” that is information about the design.
Listen and follow up
Listening is the main skill. Pay attention to what is said and to what is not. Hesitations, jokes, sighs and workarounds are clues. When someone mentions something interesting, follow it with “Can you say more about that?” or “Why is that?” The follow-up often yields more than the prepared question.
Pay attention to emotion. A passing mention of “I just gave up” or “I always have to ask my son” points toward a real need. This is also where you may notice needs related to accessible design that you had not considered.
Take notes and look for patterns
Write down quotes and observations as plainly as you can, separating what people said from your interpretation. Soon after each conversation, add your impressions while they are fresh.
After several interviews, look for repeated points: things that came up more than once, workarounds that several people use, tasks that cause trouble. Group similar notes together. Be careful about the loudest voice and the single memorable story. Counting is useful, but a very small number of interviews cannot give reliable percentages, so avoid talking as if you had measured the whole population.
Then go back to your problem statement. Does it still describe the real need? Revise it.
A short sample plan
- Introduce yourself and explain that you are learning, not testing the person.
- Ask about their background with the activity in general.
- Ask for a recent example in detail.
- Ask what is hard, annoying or time-consuming.
- Ask what they do about it now.
- If useful, show a rough sketch or model and ask for reactions, after the earlier questions are done.
- Thank them and ask if you may follow up.
Interviews and testing are different
Interviews help you understand needs. They are not good for judging whether a particular design works, since people are poor at predicting their own behavior. For that, watch them try a model, which is the subject of usability testing. The two methods work well together: interviews early, observation of prototypes later.
The goal is not perfection, only less guesswork. Even a few honest conversations usually teach a maker something unexpected.