How many customers are reliable enough for interviews?
The key point is when you stop hearing new information, not a fixed number of interviews.
For in-depth qualitative interviews, a common reference point is about 10 to 15 sessions for each customer group. This is the saturation point, meaning the conversations no longer yield new insights. Fewer than 5 sessions can lead to hasty conclusions based on a few isolated cases. After 6 to 7 sessions, the main patterns usually begin to emerge. The real principle is to stop when you stop hearing new information, not when you reach a fixed number.
- a diverse group, needs to be certain before major decisions, targeting 10-15 interviews
- a simple group, after 6-7 interviews, responses have repeated
- need to measure accurate rates, then switch to quantitative surveys
- Conclusions from two or three meetings are not enough to see the patterns.
- use a few qualitative interviews to claim percentage rates
You ask for numbers, but the real answer is a principle: stop interviewing when you no longer hear something new, not when you hit a fixed target. A benchmark of 10 to 15 interviews per group is common practice, not a hard scientific rule. Understanding why that number exists will help you know when you can stop earlier and when you need to go further.
What is saturation and why is it more important than the number
The saturation point is when new interviews no longer provide new insights. The eighth customer says what the sixth customer said. The twelfth customer confirms what you have already seen, without adding new perspectives. This is a sign that you have covered the main patterns in that group.
According to the research by Guest, Bunce, and Johnson (2006), after about 6 to 7 interviews, the main themes usually begin to emerge. With a relatively homogeneous group, most core patterns typically appear within about 10 to 15 sessions. However, this is a reference point, not a guarantee. The more diverse the customer group, the later the saturation point will be reached.
Deciding factors for whether you need more or less sessions
- Diversity of the customer group: customers using the product in one way, in one context will reach saturation quickly. Customers with many different usage situations and reasons for purchase need more sessions to cover everything.
- The quality of each interview: ask deeply, uncover the real motivations; one good session is worth three superficial ones. Running 15 sessions but only scratching the surface does not achieve true saturation.
- Number of customer groups you are researching: This is the most critical point. The number of 10 to 15 is counted by group, not in total. Two different customer groups mean two separate research efforts.
- Decision risk level: Small decisions, quick tests, can accept fewer failures. Major decisions about positioning or product categories should be thoroughly explored to ensure certainty.
When to stop early, when to go further
Need more than 15 sessions when: the customer file is complex, with clearly different usage situations, or after 12 to 13 sessions, new and surprising insights still emerge. This is a sign that the group is not homogeneous enough or you have not reached the right segment.
Common mistakes when counting interviews instead of listening for saturation
- Conclusions from 2 to 3 sessions: too few to distinguish true patterns from random cases. This is a common and dangerous mistake because interviewers often feel they "understand already" after just a few initial sessions.
- Combining different groups into a common sample: when combined, the smaller group is drowned out by the larger group. You leave with the understanding of the majority and completely overlook the minority group that may be key customers.
- Running enough sessions but asking superficially: the numbers look sufficient on paper but do not dig deep into the real reasons. A customer says, "I like it because of the good quality," but you do not follow up with, "What does good quality mean to you specifically?" That session does not count as in-depth.
- Using qualitative interviews to declare percentage rates: "8 out of 10 customers say they are willing to pay more" from 10 interviews has no statistical value. Qualitative is to understand why, quantitative is to measure how much.
These two methods cannot replace each other. Using the wrong method for the wrong goal may yield seemingly solid results, but it won't hold up when making real decisions. With limited budget and time, focus on the quality of each session, ask the right group, and closely monitor whether the sessions are repeating. Five in-depth interviews with the right group provide more insight than fifteen superficial ones.
Decision checklist
Topic: How many customers need to be interviewed to gain reliable insights. Sinh Vũ guide, sinhvu.com
Select each item you find appropriate, then print or save as PDF to take with you.
If you have marked most of the signs above, this is the time to discuss in more detail. Sinh Vũ can help you review and propose a direction.
References
Guest, Bunce, and Johnson (2006), How Many Interviews Are Enough. Clayton Christensen, Jobs to be Done (HBR). Practical experience in the strategic process and brand positioning of Sinh Vũ.
Frequently asked questions
I talked to 3 clients and they all said the same thing, is that enough?
Not enough to draw a solid conclusion. Three sessions are insufficient to determine whether the repetition is a true pattern or random due to your selection of similar samples. You should continue until at least 6 to 7 sessions in that group to see if the answers are truly stable. Only if nothing new emerges afterward can you have a basis for a tentative conclusion.
Can qualitative interviews tell me what percentage of customers think that way?
No. Qualitative interviews are used to understand why customers act or think that way, not to measure precise ratios. If you need reliable percentage figures, that is the job of quantitative surveys with a much larger sample size. Using a few interviews to infer ratios is a common methodological error and can easily lead to wrong decisions.
If I have multiple different customer groups, how should I approach that?
Billing by group, not by total. If you have two different customer segments, such as individual clients and corporate clients, each group needs its own separate meetings. Combining them and counting the total will overlook smaller groups and blur important differences between them.
This article is for reference. The scope, pricing, and specific commitments of Sinh Vũ are detailed in the proposal and signed contract.