An in-depth interview (IDI) is a one-to-one conversation between a researcher and a participant, usually lasting 45 to 90 minutes, that explores a subject in detail: what someone does, why they do it and how they feel about it. It's the method to reach for when you need depth from each person, privacy for a personal subject, or one person's decisions without a group around them.
IDIs are one of the oldest tools in qualitative research, and still one of the most used. This guide covers what they are, when to choose one over a focus group, how many you need, and how to run them well, online or in person.
What is an in-depth interview?
An IDI is a conversation, not a questionnaire read aloud. The researcher works from a discussion guide (a plan of topics and questions in a sensible order) but follows the participant's lead, asking follow-up questions, or probes, wherever an answer opens something up. That flexibility is the point: you find out things you didn't know to ask about.
Most IDIs are semi-structured. Every participant covers the same ground, so you can compare across interviews, but the order and the follow-ups change from person to person. They can take place by video, by phone or face to face, at a desk, in someone's home or alongside them in a shop.
How long should an IDI be?
For most consumer topics, 45 to 60 minutes is enough to build rapport, cover the guide and probe properly. Complex subjects, business-to-business research and interviews with experts often run to 75 or 90 minutes. Much beyond that and attention drops, particularly on video, so a long interview is usually better split into two sessions or supported by a short task beforehand.
Interviews shorter than half an hour can work for a focused reaction to a single idea, but they tend to become structured interviews: useful, but without the depth that gives the method its name.
Should you run IDIs or a focus group?
Both are valuable, and many projects use both. The choice comes down to whether you need each person's own story or the conversation between people.
Choose IDIs when
You need each person's full story or decision process. The subject is personal, or people might not speak freely in front of others. Participants are experts, senior or competitors who can't share a room. You want detailed reactions to stimulus without one view swaying the rest.
Choose a focus group when
The interaction is the insight: people building on each other's ideas, reacting to each other, or using the language they'd use with peers. You want a quick read across a range of views in one session, or you're exploring how a group negotiates an opinion.
The evidence is more balanced than the usual rules of thumb suggest. In a randomised study, Guest, Namey and colleagues (2017) asked the same questions in focus groups and individual interviews. Person for person, the interviews surfaced a wider range of ideas. But some sensitive, personal disclosures came up more readily in the groups, which is a good reason not to assume a sensitive subject always means interviews.
There are options in between, too. Paired interviews (dyads) and triads, with two or three people, suit couples, friends or colleagues who make decisions together.
How many IDIs do you need?
It depends on how varied the people are and how many distinct groups you need to hear from. The usual measure is saturation: the point at which new interviews stop adding new themes.
That figure, from Guest, Bunce and Johnson (2006), is the source of the familiar "12 interviews" rule. But their sample was quite uniform. When Hagaman and Wutich (2017) looked at research across four sites and cultures, 16 or fewer interviews were enough within each site, but finding the themes that ran across all of them took 20 to 40.
In practice, 8 to 12 interviews per distinct group is a sensible starting point, with more when people differ a lot or the project spans countries. There's more on this in what decides your sample size.
How do you run IDIs online?
Video interviews are now the default for most projects. You can reach people anywhere, fit around their day, and record everything without anyone in the room. What makes them work well is mostly preparation:
- Make joining easy. Participants should join from a link in their web browser, with nothing to install, and check their camera and connection well before the interview, not five minutes before.
- Keep the guide and stimulus to hand. Have images, videos and concepts ready to show, so the conversation doesn't stall while you look for a file or share your screen.
- Leave room for silence. On video it's tempting to fill every pause, but a few seconds of quiet is often when the most considered answer arrives.
- Note the moments that matter as they happen. A quick mark and a word or two at the time saves hours of searching the recording later.
Can clients watch an IDI?
Yes, and it's one of the best ways to bring a client close to the research. Observers watch from a separate backroom that the participant can't see, so the interview stays a one-to-one conversation. They can talk among themselves and pass the researcher a question to ask at the right moment, rather than interrupting. Our guide on running online focus groups covers the backroom in more detail, and it works the same way for interviews.
Can you run IDIs in other languages?
Yes. The simplest route is a moderator who speaks the participant's language. Where that isn't possible, an interpreter can join on a separate line that observers listen to, so the participant hears only the moderator and the client follows in their own language. Recordings can then be transcribed and translated, so every interview can be analysed together.
What about in-person interviews?
Some interviews belong in the room: in someone's home, in a shop, or anywhere the setting is part of the story. Recording them on a phone or tablet gives you the same film and transcript as an online interview, so in-person and online interviews can be analysed side by side, and clients can still follow along remotely.
How do you analyse in-depth interviews?
Analysis starts with the transcript. The researcher reads each interview, notes what stands out, and groups those observations into themes across the whole set, using participants' own words as evidence.
AI now does much of the first pass well: transcribing, translating, summarising each interview and pulling out key points with the quotes behind them. That leaves more time for the part that needs a researcher, deciding what the findings mean and what the client should do about them. Whatever tool you use, check what the AI gives you against the transcript, and treat it as a draft, not a conclusion.
For the debrief, short video clips of participants saying it in their own words are often more persuasive than any slide.
Should you combine IDIs with other methods?
Often, yes. A short online task before the interviews, such as photos of where someone uses a product or a few days of diary entries, means you start each interview already knowing the person, and spend the time on why rather than what. A follow-up task afterwards lets people add what they thought of later. Running both in the same project keeps everything together for analysis. There's more on this in asynchronous qualitative research explained.
The bottom line
In-depth interviews give you one person's full story, in their own words, with the chance to follow it wherever it leads. Choose them when individual depth matters more than group interaction, plan for 8 to 12 per group as a starting point, and make the most of the time by preparing well and letting AI take the routine work off your hands. Qualzy Live is how we run them, online and in person.
