Good interviews start before the call
The best interviews start before the call. A clear research question, explicit assumptions you're testing, and a script you've thought through. When you know exactly what you're trying to learn, the conversation flows naturally. You know when to probe deeper and you recognize a useful answer when you hear one.
Recording the conversation
Record if you can. You'll miss details if you're only taking notes, and you can't maintain eye contact while writing. Ask permission at the start; most people are fine with it once you explain it's just for your notes.
That said, always take some notes as backup. Recordings fail. Batteries die. And the act of writing helps you stay engaged.
The first two minutes
Start with a minute of small talk. Weather, their location, how their day is going. Nothing profound. This isn't wasted time; it helps people relax and shifts them out of "being evaluated" mode.
Then set expectations: how long the conversation will take, what you're trying to learn, and that there are no wrong answers. Keep it natural. Something like: "I'm trying to understand how people handle [topic]. There's no right or wrong here, I'm just curious about your experience."
This matters more than it sounds. People open up when they know what to expect.
One question at a time
It's tempting to stack questions: "What do you do when X happens, and how often does that come up, and is that frustrating?" Don't. People will answer the last question and forget the rest.
Ask one thing, wait for the full answer, then ask the next. It feels slower, but you get better responses.
Follow-ups that work
The scripted questions get you started, but follow-ups are where the real insights come from. A few that work well:
"Tell me more about that." Simple, but effective. It signals genuine interest and gives people permission to elaborate.
"What happened next?" Keeps the story going. Often the most useful part of an answer comes after the initial response.
"Why was that?" Gets at motivation. Use sparingly. Too many "why" questions can feel like an interrogation.
Mirroring. Repeat the last few words they said, with a questioning tone. "It felt frustrating?" This invites them to continue without steering them.
"Can you give me an example?" Moves from abstract to concrete. When someone says "it's always a hassle," asking for a specific instance gives you the story behind the statement.
Don't judge the setting
I once had a participant join from his car, phone propped on the dashboard, still in a parking lot. I didn't expect much. But once I started asking questions, he opened up completely: thoughtful, specific, engaged. One of the most valuable interviews in that project.
The setting doesn't determine the quality of the conversation. Someone in a messy home office might give you gold. Someone in a perfect meeting room might give you nothing but polished non-answers.
Reading between the lines
Some participants want to be helpful. Too helpful. They sense what you're hoping to hear and give you exactly that. Over time, you learn to notice:
Answers that come too fast. Genuine reflection takes a moment. Instant agreement often means they're telling you what you want to hear.
"I think I would..." This signals speculation, not experience. Gently redirect: "Has something like that actually happened?"
Enthusiasm about features vs. problems. If they're excited about your proposed solution but vague about the problem it solves, be skeptical.
What they don't mention. If you expected a topic to come up and it didn't, that's data too. Either it's not as important as you thought, or they're avoiding it.
There's no checklist for this. It's pattern recognition you develop over time. Sometimes you have to trust your gut and set aside data that doesn't ring true.
Avoid hypotheticals
"Would you use this feature?" "Would you pay for that?" These questions feel natural, but they're almost useless. People can't reliably predict their own future behavior. They'll say yes to be polite, or because it sounds good in the abstract.
Even with a well-prepared script, a hypothetical can slip out mid-conversation. When you hear yourself ask "would you...", just rephrase on the spot: "Actually, let me ask it differently. Have you ever looked for something like this?" or "What do you do now when this happens?"
Past behavior is the best predictor of future behavior, and it's much harder for people to embellish what they've actually done.
Making space for silence
Silences are where the good stuff often emerges. People need time to think. If you jump in too quickly, you get the first thing that comes to mind instead of the insight that actually matters.
My approach: I tell participants at the start that I'll sometimes leave pauses on purpose, not to make things awkward, but to give them space. Once it's announced, the silences feel natural.
Checking you're solving the right problem
On a project in the equipment rental industry, the brief was clear: help rental companies offer smaller time slots and automated payments. That's what we asked about. But we also explored the surrounding workflow, what happens before and after.
What we found surprised us. Nobody cared much about time slots or automated billing. What came up repeatedly was identification: figuring out who they were renting to, whether that person was trustworthy. That was the real daily friction.
The project took a completely different direction. If we'd only asked about the original brief, we would have missed it entirely. Exploring the context around your research question, not just the question itself, often reveals what actually needs solving.
A question worth asking at the end
At the end of every interview, ask: "Is there anything I should have asked but didn't?"
This simple question often surfaces insights you wouldn't have found otherwise. People know their domain better than you do. Give them a chance to tell you what you missed.
It gets smoother
The first few interviews in a project can feel clunky. You're still finding your rhythm with the script, learning which questions land and which need rephrasing. That's normal and expected.
What changes with experience isn't that you stop making mistakes. It's that you recover faster. You notice when something isn't working and try a different angle. You get comfortable with the imperfections and stay focused on the goal: learning something useful from each conversation.
From conversation to notes
Right after the call, take five minutes to capture what stood out. Don't wait; details fade fast.
What to note:
- Key insights. The core of what they said, in your own words. The recording has the exact quotes; you'll pull those later.
- Your reactions. What surprised you? What confirmed what you expected? What felt off?
- Gaps. What did you wish you'd followed up on? What did they mention that you didn't explore?
This isn't analysis. That comes later. It's just getting the raw material down before you forget it.