So you’ve convinced your organization, team, or client that you need to talk to users.
Not just usability tests (also important!), but sessions that you’ll conduct at the beginning of your discovery process to help you: explore problems and hypotheses, invalidate assumptions, develop personas, test early prototypes, and figure out the “why”s behind the “what.”
This is great news.
Except: you don’t have a lot of research chops on your team. Or, you’re the only one with research chops on your team. Or, even though your product folks and designers are on board to pitch in, everyone has a little trepidation about putting their researcher hat on.
No one is going to become the perfect moderator overnight. In the weeks and months ahead there will be cringe-worthy leading questions, awkward silences, and dozens of precious “why”s left unspoken.
But this isn’t life or death, and some learnings are always better than none. With practice, repetition, and a few best practices, your product team can transform into a team of the best user researchers ever.
Formulating useful questions
The newer you are to research, the more important it is to come prepared with well-crafted questions about the topics you want to explore. It may feel a little stiff reading them off at first, but it’s better than trying to make it feel organic and ending up with biased findings.
- Avoid leading or closed questions
This falls into the “minute to learn, lifetime to master” advice pile. It’s essential not to bias the interview by embedding your own assumptions into your questions, but it’s easy to do it unconsciously– especially if the answer seems obvious, or you’re on the spot and didn’t formulate a non-leading question beforehand.
Not useful → Would you say ellipticals are better than treadmills?
Better→ What type of exercise gives you the best workout?
Not useful → Do you like to use felt tip pens?
Better → What types of tools do you use to draw?
Not useful → So as a designer, you use photoshop a lot, right?
Better → In your survey you mentioned you work at Company X. What does your role involve?
2. Probe behavior, not intention
Good research is about understanding what your users do today. Leave wants, desires, and intentions to your marketing team.
Not useful → How many times do you plan to go to the gym?
Better → How many times have you been to the gym in the last two weeks?
3. Focus on specifics
Not useful → How do you usually…?
Better → Walk me through the last time you…
Better → What were the steps you took to…and then what did you do?
Better → Can you give me an example? (After they mention something)
Not useful → Is that normal?
Better → How many times did that happen last week?
Better → Do your coworkers also do it that way?
A conversation that isn’t
Ideally, an interviews should feel like a natural conversation to your user (Not an interrogation with a question-robot). But a lot of the techniques that make for a good conversation can actually make for a bad interview. Even though you’re trying to keep it natural, you have to unlearn some of your best conversation skills to help prevent biasing the session.
- Tempting as it may be, avoid agreeing or disagreeing with the user, talking about yourself (or your friend, or co-worker…), or explaining the way the product should be or what your team’s intention was
- Be careful about paraphrasing their statements, finishing their sentences, or putting words in their mouth. Your helpfulness may be biasing or twisting their answers.
- A little vagueness is OK. Your subject shouldn’t necessarily be able to figure out exactly what you’re working on or what your assumptions are by the end of the interview (again, a sign that you may be leading them)
Setting the stage properly
In addition to thanking your interviewee for their time, breaking the ice with a little small talk, confirming they’ve signed the NDA, and getting permission to record…
- Give some context: “We here to learn about you and people like you as it relates to XYZ”
- But not too much: “We’re having problems getting customers to convert on the checkout flow”
- Encourage honesty. Let them know you didn’t design what they’re looking at (lie if necessary), so they can be honest (“you can’t hurt my feelings.”)
Try to shut up
Use awkward silences to your advantage. The juiciest insights often happen in the spaces between all those well-crafted questions you prepared. Your interviewee wants to fill the silence just as much as you do, so let them!
- Pause after you say something
- Pause after they say something
- Don’t jump to the next question
- Don’t interrupt (including acknowledgements)
- “Let people speak in paragraphs” — Steve Portigal
Beware of the query effect
“People can make up an opinion about anything, and they’ll do so if asked. Users can comment at great length about something that doesn’t matter to them, and which they wouldn’t have given a second thought of if left to their own devices.” — Jakob Nielsen, Interviewing Users
Not leading users isn’t just about asking open-ended questions or putting words in a user’s mouth. One of the most tempting and dangerous ways to get responses that belong in the discard pile is to ask about something that didn’t naturally come up — like that filter they overlooked in your wireframe, or their thoughts on how best to share an article on their social networks (when that’s not something they do today).
So how do you get feedback on your burning, must-answer questions without bringing them up out of the blue?
- Consider if there are broader, non-leading questions you can ask up front that will more likely take the conversation in the direction you’re curious about. (“Describe a positive experience you’ve had reading an article on that website.”)
- Create goal-based scenarios when showing a prototype or design. (“Let’s go back to that example where you were searching for happy sloths. Imagine if this were the page you landed on. What would you do next to find the image you wanted?”
- Have a list of your “must-answers” handy during your interview and check them off as they come up (organically). Towards the end of the interview, take a look at that list and choose ONE or TWO topics to broach. Ask about them in as non-leading a way as possible (write them out beforehand to prevent bias as much as possible). Take their answers with giant grains of salt and many asterisks.
Would you use this?
This is an extremely tempting question to ask (especially after your interviewee has reacted positively to something you’ve said or shown them.) But asking this questions is going to give you false positives that can end up mis-informing your design process.
Being tough on your assumptions now will spare you from headaches later.
Some questions you can ask instead:
- Why is it “cool”?
- How is it “helpful”?
- How would this change what you do today?
- Can you give me an example of how you would use this?
- Going back to that time you searched for happy sloths, how would this have fit into that process?
- When wouldn’t you use this?
- Is this more helpful than X?
- If you could only have 3, would this be one of them? (Why?)
Regrets, you’ll have a few
You can’t ask a recording what they meant later, which is why it’s so important to clarify and dig into what your interviewee is saying as it’s happening. If you’re unsure of what they said or why they said it, you’re going to end up with cryptic, half-usable information at the end of your session.
Here are some good places to start when it comes to clarifying and probing during an interview:
- Repeat what they said (verbatim) and pause
- “Tell me more about that…”
- “Can you give me an example?”
- “What do you mean by that?”
- “What do you mean by thing they said?”
- “What did you expect to happen?
- “Why do you call it ‘the Hellmouth’?”
- “How do those departments work together?
- “When you say “her,” who exactly are you referring to?”
There’s no such thing as an obvious answer! It is always safer to ask “why,” even if you think you know the answer. Getting it in their words will be far more valuable.
Paraphrasing is the gateway to leading
Interviewee: “I often browse through 30 or 40 pages of results before I find what I’m looking for.”
Researcher: “So searching is pretty time consuming for you.”
Interviewee: “Um… yeah, I guess so.”
Paraphrasing: great for conversations, bad for interviews. Why? Because you’re essentially putting words in your subject’s mouth — words that are embedded with your own assumptions. Unless you get it totally wrong, your user probably won’t correct you (too much effort, you were close enough…). But just because they don’t correct you doesn’t mean you got the whole story. Instead of paraphrasing, try clarifying questions that are less likely to bias your interview:
- “Why do you do that?”
- “How do you feel about that?”
- “How often does that happen?”
- “When you say XYZ, what do you mean by that?”
You are the best researcher ever.
Confidence is key! If you’re unsure during the interview, your user will be too. Start strong, end strong, and speak clearly and confidently throughout (even if you don’t feel that way yet , you will soon!).
A few things to keep in mind:
- Trust the question
- Don’t trail off
- You can be open-ended and direct at the same time
- It’s ok to pause or regroup (the silence isn’t as long as it feels)
- They have no idea how the interview is going
You will get better at this
It’s hard, but try to force yourself to listen to your interview recordings so you can reflect on what you’re doing well and what you can get better at. Don’t be scared to ask your teammates for their honest impressions and feedback, too (they’ll especially be able to notice things like your tone and body language).
When it comes down to it, user research is all about practice, preparation, and asking why. Your sessions may feel awkward or clumsy at first, but they’ll still be invaluable to your team.
Want to share with your team? Get all the tips and tricks from this post (plus more!) in this handy deck:
Talisa Chang is a interdisciplinary product and UX consultant who likes to help teams learn before they build.