FGI · Customer Interview
Focus Group Interview and Customer Interview
What is an FGI?
- Surveys tell you how many; interviews tell you why
- Ask about past behavior, not future intent
- Skews results if you only talk to happy customers or generalize from the numbers
A customer interview is not a session for asking 'What do you think of our product?' It is research that listens to 'why they did it' in the customer’s real experience, which survey numbers cannot tell you. A survey tells you what percentage of customers are dissatisfied; an interview tells you in what situations that dissatisfaction arises. That is why the order matters. When you don’t know what to ask, start with interviews; when you need to confirm how many customers feel that way, follow up with a survey.
There are two main formats. A focus group (FGI) brings together several participants with similar profiles and lets them react to one another, which is a fast way to collect a range of opinions and phrasing. A 1:1 in-depth interview follows one person's experience from start to finish, so it suits purchase journeys and sensitive topics. In B2B, several people are involved in each buying decision and it is hard to put competitors' employees in the same room, so 1:1 interviews are used far more often in practice.
The key to doing it well is the questions. Hypothetical questions like 'Would you use this feature if we added it?' usually get a polite 'yes.' Ask about past behavior, not future intent. 'Tell me about the last time you ran into this problem' and 'How did you deal with it?' are good questions. With 5–8 customers of the same type, recurring stories start to emerge. Split the roles of moderator and note-taker, and write up what you heard, in the participants' own words, on the day of the interview.
Where it goes wrong in practice is hearing only what you want to hear. If sales reps recommend only customers they’re close to, you collect nothing but praise, so you must include customers who left and prospects who never signed. In an FGI, one loud voice often steers the whole discussion, or others go quiet in front of a senior participant. Also, never turn what eight people said into a statistic like '70% of customers want this.' Qualitative research is a tool for forming hypotheses, not for proving proportions.
John KimInstructor John Kim, author of B2B Marketing Design and CEO of Nanum Consulting
Seungtae ParkAI transformation practitioner who teaches AI business development and marketing automation by building the tools himself| Category | Focus group (FGI) | 1:1 in-depth interview |
|---|---|---|
| Size and duration | About 6–8 people, 90–120 minutes | 1 person, 40–60 minutes |
| Strengths | Quickly gathers a range of opinions and customers’ own words | Follow one person's experience and reasons all the way through |
| Weaknesses | Easily swayed by loud voices and seniority | Takes a lot of time per participant |
| Best-suited topics | Reactions to concepts, messages, and drafts | Purchase process, reasons for leaving, sensitive internal circumstances |
| B2B considerations | Hard to seat competitors together; scheduling is a burden | You need to meet each role involved in the purchase separately |
The method is commonly traced to “The Focused Interview,” a paper sociologists Robert Merton and Patricia Kendall published in the American Journal of Sociology in 1946. The group interview technique used in studies of listener reactions to radio broadcasts later spread to marketing research and became today’s focus group.
Consider a hypothetical case. Company A supplies appointment management software to small and mid-sized hospitals, with 420 hospital clients and 75 employees. Its year-end satisfaction survey showed a healthy overall score of 4.1 out of 5, yet 58 customers did not renew that year, double the previous year’s number.
- The marketing manager proposed interviews, saying, "The scores are good, but the survey doesn’t tell us why customers leave." The CEO approved the month of December and a KRW 3 million (≈ US$2,240) budget for participant fees.
- Participants were selected directly, not from sales team referrals. We picked 10 administrative managers from 4 renewing customers and 6 churned customers, and met each one 1:1 for 40 minutes.
- The questions focused on past behavior. When asked, "Tell me about the last time a booking went wrong," 5 of the 6 churned customers described the Monday morning right after a holiday weekend.
- In the transcript the note-taker compiled that same day, 'We couldn’t change them in bulk, so we called one by one' came up five times. The head of development said, "That never once appeared on the request board."
- The team checked this hypothesis with a survey of all customers and found that 61% of 190 respondents had the same problem. After adding the feature, non-renewals the following year fell from 58 to 27.