5 Interviews vs 20: How Many Do You Really Need?

“How many interviews do I actually need — 5 or 20?” is a question with no single right answer, but the two ends of that range represent genuinely different situations, and understanding when each is appropriate can save you either wasted time or a false sense of confidence. Here’s how to think about it.

What 5 Interviews Can and Can’t Tell You

Five interviews can be genuinely useful, particularly early in discovery, to sanity-check an initial hypothesis or explore a topic you don’t yet understand well. If four or five people, independently, describe a strikingly similar problem in similar terms, that’s a meaningful early signal worth pursuing further.

What five interviews generally can’t do is give you confidence that the pattern you’re seeing represents your broader target audience rather than a coincidence, or a quirk specific to how you happened to find those particular five people. A small sample is especially vulnerable to being skewed by recruitment source — if all five came from the same community or referral chain, the “pattern” you’re seeing might just reflect something specific to that group.

What 20 Interviews Can and Can’t Tell You

Twenty interviews, especially if drawn from a mix of recruitment sources, give you much more confidence that an observed pattern is real and not a fluke of a small or narrow sample. They also give you more room to notice sub-patterns — differences between segments of your audience that a smaller sample might not reveal.

The tradeoff is time and effort. Twenty genuinely useful interviews, each with real follow-up and depth, represent a significant investment, and if a clear pattern already emerged after ten, the marginal value of interviews eleven through twenty may be limited — see how many customer interviews before building an MVP for more on this diminishing-returns dynamic.

A Comparison

Factor	5 Interviews	20 Interviews
Time investment	Low	Significant
Confidence in pattern	Limited, vulnerable to small-sample bias	Higher, especially with diverse sources
Best used for	Early hypothesis check, narrow/familiar audience	Broader or unfamiliar audience, higher-stakes decision
Risk if used alone	Mistaking coincidence for pattern	Diminishing returns past the point of pattern clarity

Factors That Push You Toward Fewer Interviews

• You already have significant domain expertise in the space.
• Your target audience is narrow and well-understood.
• Early interviews are producing remarkably consistent, detailed answers.
• The cost of being wrong is relatively low (a very cheap MVP slice, or a low-stakes test).

Factors That Push You Toward More Interviews

• You’re entering an unfamiliar market or audience.
• Early interviews are inconsistent or contradictory.
• The decision at stake (a significant development budget) carries real cost if wrong.
• You need to understand multiple distinct audience segments separately.

The Real Answer: Watch for the Pattern, Not the Count

Rather than committing to a fixed number in advance, use the number as a rough planning guide and let the actual evidence tell you when to stop. If a clear, consistent pattern emerges by interview eight, continuing to twenty mostly adds confirmation rather than new insight. If interview fifteen still isn’t revealing a clear pattern, that itself is useful information — it may mean the audience needs to be narrowed rather than that more of the same broad interviewing will eventually produce clarity.

A Practical Approach

Start with a target range — say, 10 to 15 — as your default plan. Check in after each batch of five: is a pattern emerging, staying consistent, or still scattered? Adjust your target based on what you’re actually seeing, rather than treating your original number as fixed. This flexible approach captures the benefit of both ends of the 5-vs-20 debate without over- or under-investing in discovery relative to what the evidence actually needs.

From the Right Number to a Confident Decision

Whether it takes five interviews or twenty, the goal is the same: a clear enough pattern, drawn from a genuinely representative sample, to make a confident decision about what to build. Getting there efficiently matters more than hitting any particular number along the way.

Frequently Asked Questions

Is 5 customer interviews ever enough before building an MVP?

It can be, for a narrow, well-understood audience where the founder already has significant domain knowledge, or when the interviews are unusually consistent and detailed. For most new, unfamiliar audiences, 5 is often too few to distinguish a real pattern from coincidence.

Is 20 interviews always better than 5?

Not necessarily. If the first 8-10 interviews already show a clear, repeating pattern, the additional interviews up to 20 may add confirmation but little new insight, and the extra time might be better spent testing the idea a different way.

What determines whether 5 or 20 is the right number for my situation?

How specific and consistent your target audience is, how quickly patterns emerge, and how much domain knowledge you already bring to the process. Broader or less familiar audiences generally require more interviews to reach a reliable pattern.

Should I set a fixed number of interviews in advance?

It's reasonable to set a rough target as a planning tool, but treat it as flexible — the actual right number depends on when a clear pattern emerges or clearly fails to, not on hitting a predetermined count.

What's the risk of stopping too early, like after only 5 interviews?

With very few interviews, it's easy to mistake one or two people's specific situation for a broader trend, especially if those few people happen to share something unusual in common that sets them apart from your real target audience.