How Many Customer Interviews Do You Really Need?
Six to eight customers per round is enough for innovation research. After six to eight conversations with one group, the same themes start to repeat.
Patterns, not percentages
Qualitative research looks for patterns. It tells you what people value and why. Measuring how many people feel a certain way is a different job.
Percentages need large samples. Patterns don't. You are listening for repeated themes, reasons and language.
Mixing the two causes confusion. A board may ask why only eight people were involved. The answer is that the round was built to find something worth measuring.
Why six to eight customers per round is enough
Themes repeat quickly within one group. By the sixth to eighth conversation, you hear the same needs and frustrations again. Extra interviews add little new.
Smaller rounds keep you moving. Each round takes about a week. That pace lets you change the concept and test again while the learning is fresh.
Quality matters more than volume. Rich conversations with the right people beat thin ones with many. Recording customers' words verbatim keeps the evidence strong.
Showing beats asking in every conversation. Put rough ideas in front of each customer and watch their reactions. Then ask why. A handful of these conversations tells you more than a long list of direct questions.
One audience at a time
Test with one similar group per round. A homogeneous group gives a clear signal. Mixed groups blur it.
Use the first group as a baseline. Read the next audience against it. Differences then stand out clearly.
This scales across markets. In one multi-country TV proposition across several African countries, we studied one country in depth first. We then read the others against that baseline.
Their language reveals segments. The words each group uses show where they differ. That feeds straight into marketing.
How many customer interviews across an engagement
Expect 20 to 40 customers across a full engagement. That comes from several rounds of six to eight. Each round builds on the last.
A typical rhythm looks like this:
- Week one: set up the questions, audiences and concepts.
- Rounds one to four: six to eight customers each, one week per round.
- Each week: a workshop where your team hears what customers said.
Fewer people, more rounds. Repeated small rounds teach you more than one large round. You can act on each one before the next begins.
Each round should change something. Use what you heard to adjust the concept before the next week. A round that changes nothing was probably too safe.
Add a survey to size what you find
Use a survey once you know what to measure. Surveys size what you already know. Qualitative rounds find what's worth sizing.
The order matters. A survey written too early asks about the options you already had. It can miss the need customers would have revealed in conversation.
In one home proposition for a leading European media company, a survey pointed to home security. When customers talked about their homes, water damage and the health of the home came through. That proposition launched.
Make the numbers credible to your board. Your board won't back instinct. It will back customers.
- Show the patterns in customers' own words. Verbatim quotes make findings hard to dismiss.
- Keep the outliers. The quiet voice is often where new ideas start.
- Bring decision-makers into the room. People back what they've heard themselves.
- Pair depth with scale. Small rounds find the opportunity. A survey gives you the numbers to back it.
Start small and learn fast
The right number of customer interviews is smaller than most teams expect. Six to eight per round, one audience at a time, gives you clear patterns quickly. Twenty to forty across an engagement gives you confidence to act.
If you'd like to see how this works in practice, get in touch. We can talk through a sprint shaped around your audiences.
Read next: concept testing four rounds · customer research for innovation faster horses · sprints