User interviews: how to find out what your users actually need

Most projects we see go wrong start with a sentence like "our customers want an app". The owner is sure, the sales director is sure, and the budget is set. Then someone talks to six customers, and it turns out they don’t miss an app at all. They miss being able to see the status of their order without phoning. That is a completely different and often far cheaper solution. User interviews are the fastest way to find out this kind of thing before the money is spent. This guide shows you how to do it yourself, and when it pays to get help.

What is a user interview, and when should you use one?

A user interview is a structured conversation with one person from your target group. The aim is to understand how that person handles a particular task today, what is awkward about it, and what it costs them in time, money or frustration. You ask open questions about concrete situations and listen more than you talk. The method is strongest early in a project, when you are choosing a direction, and again when you are weighing a major new feature. It fits well with a product discovery workshop, where the findings become a prioritised plan.

Interviews and tests are often confused. In short, the interview tells you which problem is worth solving, and the test tells you whether your solution works. We have a separate guide to running a five-person usability test, which is the natural next step once you have a prototype.

How many user interviews do you need?

Five to eight interviews per user group is a good starting point. After four or five conversations the same stories start to repeat, and that is exactly what you are looking for: patterns that hold across people. The key word is user group. If your system serves customers, staff in the field and an administrator in the office, those are three groups with three different working days. Then you are better off with 12–15 shorter interviews than six long ones with the wrong people. Stop when the last two conversations add nothing new, and spend the time on synthesis instead.

How do you recruit the right participants?

Recruiting is the part that most often delays an interview round. Start with your own lists: customers, former customers, leads who didn’t buy, and staff who use the system every day. Write a short invitation that explains the purpose, the length and that there are no right or wrong answers. Offer a small thank-you, such as a gift card, especially to people outside your customer base. B2B users often say yes when they feel they get a say in a tool they will have to use themselves.

• Your customer list: fastest and cheapest, but pick across size, industry and how long they have been customers.
• Lost leads and former customers: the most honest answers about what is missing.
• Internal users: staff in customer service, the warehouse or the field know the daily workarounds.
• Network and LinkedIn: good for new audiences, but needs a clear screener.
• A pop-up on your website or in the app: when you want active users while the experience is fresh.
• A recruiting agency: fast and precise for consumer audiences, but the most expensive route.

Screener: five questions that pick the right people

1. What is your role in your company? (makes sure you talk to the person who actually does the task)
2. How often do you do [the task]: daily, weekly or less often?
3. Which tool do you use for it today? (spreadsheet, paper, a system, a colleague)
4. When did you last do it?
5. May we record the conversation for internal use, and does 45 minutes on video suit you?

Which questions should you ask? A question guide you can copy

1. Tell me a bit about your role and what a normal week looks like.
2. When was the last time you had to [do the task]? Take me through it from the start.
3. What did you do first? And then?
4. Which tools, files or people were involved?
5. Where did it slow down, or where did you have to do something twice?
6. What happened afterwards if something went wrong?
7. How often does that situation come up?
8. Have you tried other ways of solving it? What made you stop?
9. What does it cost you when it goes wrong, in time, money or frustration?
10. Who else is affected by it?
11. If you had a magic wand, what would you change?
12. Is there anything I should have asked, but didn’t?

Notice that nearly every question is about something that has already happened. That is deliberate. People are poor at predicting their own future behaviour and good at retelling what they did last Tuesday. "Would you use an app for that?" gets a polite yes. "What did you do last time?" gets you the real workflow with all its workarounds. The best follow-up questions are short: "Why?", "What do you mean by that?" and "Can you show me?"

Which mistakes ruin a user interview?

• Pitching: you present the solution and get polite nods instead of insight.
• Talking too much: the interviewer should speak for less than a quarter of the time.
• Filling silences: wait three extra seconds, and the important sentence often follows.
• Interviewing the wrong person: the manager who buys the system when the employee is the one using it.
• Taking notes alone: one person asks, another writes down quotes word for word.
• Dumping every finding into one document without grouping: then you only remember the last interview.

How do you turn interviews into decisions? Synthesis step by step

1. Write a short summary within an hour of each interview, while it is fresh.
2. Pull out observations as short notes: one observation or one quote per note.
3. Group the notes by theme, such as "double entry" or "no overview", on a board or in Figma/FigJam.
4. Count how many participants sit behind each theme. Three out of six is a pattern.
5. Phrase each theme as a need: "As a warehouse manager I need to see today’s orders in one place."
6. Prioritise the needs by frequency and consequence, and agree on the top three.
7. Turn those three into concrete next steps: a prototype, a user story or a survey.

What we see in practice is that synthesis is where the value is either created or lost. Many companies run good interviews and end up with a folder full of notes that nobody opens again. So set aside half a day with the people who will make the decisions and go through the themes together. Once the owner has personally heard three customers describe the same problem, the priority is rarely up for debate afterwards. The needs can then be rewritten as user stories that developers can estimate and build from.

What do user interviews cost? A worked example

With a freelance UX designer at a typical DKK 600–750 an hour, a round of six interviews therefore lands at around DKK 15,000–30,000 plus any gift cards. Set that against a real project: a 25-person company wants a new booking system for about DKK 150,000. Six interviews show that customers phone because they want a time confirmed within an hour. The solution becomes a simpler booking flow with automatic SMS confirmation, and two planned modules drop out of the first version. The saving is larger than the whole research budget. To see how research is priced alongside the rest of the design work, read what UX design costs.

Should the interview happen on video, by phone or on site?

Video is the default: it is easy to book, easy to record, and the participant can share their screen and show how they work. Phone works for busy B2B users, but you lose facial expressions and screen sharing. Visiting users on site gives the richest findings, because you see the sticky notes on the monitor, the spreadsheet next to the system and the colleague who keeps getting asked. It costs more time, but it is indispensable when the work happens in the field, in a warehouse or on a production floor. We have, for example, carried out field research and clickable prototypes for Grundfos, where the context mattered as much as the answers.

When should you run interviews yourselves, and when should you get help?

You can easily run the first round yourselves with the guide above, especially if you have a good customer list and a person who is curious and a good listener. Get help when the topic is sensitive, when you already hold a strong opinion about the answer, or when the findings need to go straight into design and development. An outsider asks the obvious questions you have stopped asking yourselves. At Ceptiv, research, design and development sit in the same team, so what we hear in the interviews goes directly into the prototype without being translated three times. Read more about our user research, or get a fixed price within 24 hours.