Qualitative Research Explained: In-Depth Interviews, Focus Groups and Finding the “Why”

What qualitative research is for

Numbers tell you what is happening. Conversations tell you why. A survey might show that 40% of customers who try your product do not buy it again. Qualitative research finds out what those customers felt, what they expected, what disappointed them and what language they use to describe it.

Qualitative research works with smaller numbers of people in much greater depth. Instead of fixed answer options, it uses open conversation, observation and creative exercises to understand motivations, emotions, context and decision-making. It is especially valuable early in a project, when you do not yet know which questions to ask, and after a survey, when you need to understand surprising results.

This guide explains the main qualitative research methods, how studies are run and how to judge whether the findings are reliable.

Qualitative or quantitative?

The two are partners, not rivals. A common sequence is qualitative research to explore and generate hypotheses, followed by quantitative research to measure how widespread each view is. Sometimes the order reverses: a survey reveals a pattern, and interviews explain it.

In-depth interviews (IDIs)

An in-depth interview is a one-to-one conversation, usually 45 to 90 minutes, between a trained moderator and a respondent. IDIs are ideal for sensitive topics, complex decisions, professional audiences and situations where group dynamics would distort answers.

They are the method of choice for B2B and healthcare research, where senior professionals are hard to bring together and speak more openly alone. Our business panel and healthcare panel guides explain recruitment for these audiences. IDIs can be conducted face to face, by video call or by telephone, depending on the audience and the need to show materials.

Focus groups

A focus group brings together six to eight people with something in common, such as category users or a particular life stage, for a guided discussion of about 90 minutes to two hours. Groups are valuable when interaction adds insight: people build on each other’s ideas, react to concepts together and reveal shared language and norms.

Groups work well for exploring categories, testing advertising and concepts, and understanding how people talk about a topic. They are less suitable for sensitive subjects or for audiences with very different levels of knowledge, where dominant voices can steer the discussion. Skilled moderation keeps every voice in the room.

Online focus groups by video are now common and work well for many audiences, particularly busy professionals and geographically spread participants. In-person groups remain valuable where physical products must be handled or tasted.

Online communities and diaries

Online qualitative communities, sometimes called bulletin boards, let 15 to 40 participants take part over several days or weeks. They answer questions, upload photos and videos, keep diaries and respond to each other, at their own pace.

This method captures behaviour over time, such as a week of cooking, a month of using a new app or a patient’s treatment journey, which a single interview cannot. It also gives participants time to reflect, often producing richer answers than a live session.

Ethnography and observation

What people say and what they do are not always the same. Ethnographic research observes people in their natural context: at home, in a shop, at work or on the move. A researcher might watch how a family plans meals, how a shopkeeper arranges stock, or how a technician uses a tool on site.

Observation often uncovers needs people cannot articulate because the behaviour has become automatic. Workarounds, frustrations and improvisations are rich sources of product and service ideas. Mobile ethnography, where participants record short videos of their own routines, extends this approach at lower cost.

Recruitment: the foundation

In qualitative research, every participant matters, so recruitment quality is critical. A study with eight interviews cannot absorb two respondents who do not genuinely fit the criteria. Careful screening checks category usage, behaviour and attitudes, and avoids “professional respondents” who take part in research too often.

Good practice includes detailed screeners, telephone re-screening before sessions, limits on how recently someone has taken part in research, and a mix within each group so that views are not artificially uniform. Our sampling & recruitment team recruits from our panels, verified networks and, where needed, door-to-door or community recruitment.

Writing a discussion guide

The discussion guide is the moderator’s map. It sets out topics, key questions, probes and exercises, with approximate timings, but it is not a script. Good guides move from broad to specific: warm-up and context, current behaviour, attitudes and needs, reactions to stimuli such as concepts or ads, and a wrap-up that captures final thoughts.

Projective techniques add depth. Asking people to imagine a brand as a person, complete a sentence, sort pictures or describe how a friend might react can surface feelings that direct questions miss.

The art of moderation

A skilled moderator creates a comfortable space, listens far more than they speak, and follows interesting threads while still covering the guide. They ask open questions (“Tell me about the last time…”) rather than leading ones (“Don’t you think…”), and they probe beneath first answers with simple prompts: “Why is that?”, “Can you give me an example?”, “What did that feel like?”.

Moderators also manage group dynamics, draw out quieter participants, gently contain dominant ones and stay neutral, never revealing what the client hopes to hear. In healthcare and other regulated sectors, they must also recognise and handle adverse events and compliance issues correctly.

Qualitative research in India

India’s diversity makes qualitative research especially rewarding and demanding. Language, region, city size, community and life stage all shape attitudes. Sessions should be conducted in the language participants are most comfortable with, by moderators who understand local context, and studies should include enough locations to capture regional differences.

Practicalities also matter: timing sessions around work and family routines, choosing comfortable venues, considering whether mixed-gender groups will allow open discussion on a topic, and providing transcripts and translations that preserve meaning rather than translating word for word.

From conversations to insight

Analysis is where qualitative research earns its value. Researchers review recordings and transcripts, code themes, compare patterns across participant types, and look for tensions and contradictions, such as when people say price matters most but describe choosing the more expensive option.

Strong findings are grounded in evidence: each insight is supported by several participants and illustrated with verbatim quotes or video clips. Equally important is honesty about limits. Qualitative findings show the range of views and the reasons behind them; they do not show how common each view is.

Reporting that leads to action

The best qualitative reports are short, visual and focused on implications. Rather than summarising every session, they answer the business questions directly: what we learned, why it matters and what to do next. Personas, journey maps, quotes and short video highlight reels bring participants’ voices into the client’s meeting room.

Many clients find a workshop more useful than a document. Walking stakeholders through the findings, discussing implications and agreeing next steps together turns insight into decisions. Our reporting & insights team designs outputs with that in mind.

An example: understanding why trials do not convert

A subscription app sees many free-trial sign-ups but few paid conversions. We recruit 16 recent trial users, half who converted and half who did not, for 45-minute video interviews, plus a one-week diary from eight new sign-ups.

The interviews reveal that non-converters never reached the feature that delivered the app’s main value, because onboarding pushed them into settings first. Diaries show the exact moments people lost interest. The recommendations are concrete: change the onboarding order, add a guided first task and adjust the day-three reminder. A follow-up survey then measures how common each issue is across all trial users.

Testing concepts and communication qualitatively

Qualitative research is especially useful for developing ideas before they are measured. Early concepts, packaging routes, advertising scripts or website designs can be shown to participants in interviews or groups to understand what is clear, what is confusing, what feels credible and what language resonates.

The goal at this stage is not to pick a winner by vote, but to understand how people interpret each idea and how it could be improved. Strong concepts often emerge by combining the best elements of several routes. Once refined, they can be tested quantitatively with a larger sample from our consumer panel.

Ethics and participant care

Qualitative research involves people sharing personal views, experiences and sometimes sensitive details. Participants should know who is conducting the research, how sessions are recorded and used, and that they can decline any question or stop at any time. Recordings and transcripts are stored securely and shared only as agreed.

Extra care applies to vulnerable participants and sensitive topics, as described in our Research Ethics policy. Fair incentives recognise participants’ time, and anonymity is protected in reports unless participants have explicitly agreed otherwise.

How many sessions are enough?

Qualitative studies are designed around saturation: the point at which new sessions stop producing genuinely new themes. For a fairly homogeneous audience, this often arrives after eight to twelve interviews. When a study covers several distinct segments, such as different regions, user types or professional roles, each segment needs enough sessions of its own.

Budget and timeline matter too, but cutting sessions too far risks missing important perspectives. A sensible plan balances coverage of key segments with depth in each, and allows a few extra sessions where early findings reveal something unexpected.

Choosing a qualitative research partner

Ask about moderators’ experience with your audience and category, how participants are recruited and verified, which languages and locations are covered, how analysis is done and what deliverables look like. Ask to see an example report and, where possible, a sample recording.

Read our complete market research guide for the wider picture, or request a proposal for your next qualitative study.

Frequently asked questions

• What deliverables do I receive from a qualitative study?

  Typically recordings, transcripts or summaries, and a findings report with implications, illustrated with quotes and, where agreed, short video highlights. Many clients also choose a debrief workshop with their team.

• How many interviews or focus groups do I need?

  It depends on how many distinct audience types you need to understand. Often 12–30 in-depth interviews or four to eight focus groups, with enough per segment to see consistent patterns.

• Can qualitative research be done online?

  Yes. Video interviews, online focus groups and online communities work well for many audiences, especially professionals and geographically spread participants.

• Do you conduct research in regional languages?

  Yes. Sessions are run in the language participants are most comfortable with, by moderators who understand local context, with transcripts and translations provided.

• Can I watch the sessions?

  Yes. Clients can observe online sessions live or watch recordings, and in-person groups can be viewed from an observation room or by live stream, with participant consent.

• How long does a qualitative study take?

  Typically three to five weeks, including recruitment, fieldwork, analysis and reporting. Smaller interview programmes can be faster.

• Is qualitative research statistically reliable?

  It is not designed to be statistically representative. It shows the range of views and the reasons behind them; a quantitative survey can then measure how common each view is.

Still curious? Ask the assistant Talk to a specialist →

Explore our Market Research services