How to Conduct an In-Depth Interview (IDI): A Practical Guide for Qualitative Researchers

There is a moment in a well-run in-depth interview where the respondent stops giving you the polite, considered answer they prepared and starts telling you what they actually think. It usually happens around the 20-minute mark, after the formal warm-up has passed and the respondent has decided you are genuinely listening.

That moment is what in-depth interviews are for. And it only happens when the researcher has done the preparatory work correctly and knows how to create the conditions for it.

What an IDI Is and When to Use It

An in-depth interview is a one-on-one qualitative research conversation, conducted by a trained moderator, that explores a participant's views, experiences, motivations, or behaviors in depth. Unlike a survey, the conversation can follow the respondent's thinking wherever it leads most productively. Unlike a focus group, there is no group dynamic to manage, no peer pressure to conform, and no dominant participant pulling the conversation.

IDIs are the right choice when: the topic is sensitive enough that group dynamics would suppress honest answers; respondents are difficult to recruit together (senior executives, specialists, people in geographically dispersed locations); you need individual-level depth rather than group-level consensus; or your research question is genuinely about the individual's experience rather than a social phenomenon.

How Many Interviews Do You Need?

This is one of the most frequently asked questions about IDI research, and the answer is not a number. It is a principle: you collect interviews until you reach thematic saturation, the point at which new interviews stop producing new themes. In practice, for consumer IDI studies, 8 to 12 respondents per distinct segment typically reaches saturation. For studies with a single homogeneous population and a focused research question, saturation can occur as early as 6 to 8 interviews.

This is why IDI samples are smaller than survey samples: each interview generates far more analytical material per respondent than a survey response does. The trade-off is depth for breadth.

Designing Your Discussion Guide

A discussion guide is not a questionnaire. It is a map, not a script. You should know it well enough to navigate the conversation without reading from it, adapting to where the respondent's most interesting answers are leading you.

Structure it in three sections

• Warm-up: general, easy questions about the respondent's background and context. These should be relevant to the topic but not yet touching the sensitive core of your research. The purpose is comfort-building, not data collection.
• Core exploration: the main questions. These should be open-ended, one question at a time, avoiding compound questions entirely. Cover the key areas of your research objectives, but do not force a linear path through them.
• Closing: a direct question about anything the respondent thinks is important that was not covered, and a clear, warm closing.

Write questions for conversation, not comprehension

Questions that begin with tell me about, walk me through, or describe for me tend to open conversations. Questions that begin with do you or is it tend to close them. Every question in the guide should be open-ended unless there is a specific reason to use a closed question.

Running the Interview

Build rapport before the guide starts

Spend the first three to five minutes before you open the guide talking to the respondent personally. Ask about their day, their role, or their environment, depending on the research context. This is not wasted time. It signals that you are genuinely interested in them as a person, which changes the quality of everything that follows.

Listen more than you talk

The single best discipline for an IDI moderator is to count how many words they are using compared to the respondent. If the ratio is not heavily skewed toward the respondent, something has gone wrong. Silence is a valid and powerful technique: after a respondent finishes an answer, waiting three seconds before speaking will often produce a more thoughtful follow-up from them than any question you could ask.

The most common mistake in an IDI is filling silence with questions. Silence is not a gap to fill. It is space for the respondent to think.

Probe deliberately, not reflexively

Probing is following up on something the respondent said to go deeper. Effective probes include: can you tell me more about that, what did you mean when you said, how did that make you feel at the time, has that always been the case?

Ineffective probes include those that telegraph the answer you are hoping for or that push the respondent toward a conclusion rather than genuinely exploring their view.

After the Interview

Write a debrief note within 30 minutes of every interview. Not a transcript. A note covering: the two or three most important things this respondent said, anything unexpected that emerged, how this interview compares to previous ones, and any analytical questions it raised.

These notes, written while memory is fresh, are often more analytically valuable than the transcripts themselves. They capture the tone, the hesitations, the things the respondent almost said. Transcripts capture words. Debrief notes capture meaning.