“We Talked to Enough People.” But Did You Really Understand Them?
Researchers often stop interviewing once answers start repeating. But hearing the same thing twice isn’t the same as understanding it. Here’s the difference, explained simply.
Imagine you’re asking people in your neighborhood why they don’t recycle. You talk to nine people. By the ninth person, everyone is basically saying the same three things. So you stop. You write up your findings and call it done.
But here’s the question nobody asks at that point: do you actually understand why people feel that way, or do you just know what they said?
Two different things people mix up all the time
In research that involves talking to people — interviews, focus groups, open-ended surveys — there’s a word researchers use a lot: saturation. It basically means “we’ve heard enough, we can stop now.” The problem is, there are actually two different kinds of “enough,” and most people treat them as if they’re the same thing.
The first kind is simple: you stop hearing new answers. Everyone starts repeating the same few points. That’s easy to notice, and it usually happens fairly quickly, often within the first several interviews.
The second kind takes longer. It’s when you actually understand why people gave those answers, deeply enough to explain it properly to someone else. Studies that have looked closely at this found it usually takes roughly twice as many interviews to get here compared to the first kind.
Why this trips people up
It’s an easy mistake to make, because the first kind of “enough” is obvious. You can literally feel it happening — people keep saying the same things, so you assume you’re done. The second kind is quieter. Nothing tells you when you’ve truly understood something deeply; you just have to keep digging until you do. So a lot of research stops at the first, easier finish line, while claiming it reached the second.
It’s not only about how many people you talk to
There’s a second thing that matters just as much: if two different people looked at the same set of interviews, would they group and label the answers the same way? A lot of the time, they wouldn’t — not because anyone did anything wrong, but because reading between the lines of what people say is genuinely a judgment call. Good researchers actually test this, comparing notes with a second person, instead of just assuming everyone would see it the same way they did.
What this means in plain terms
Don’t stop just because answers start repeating. That tells you what people are saying. It doesn’t tell you that you understand why.
Check your work with someone else. If a second person looks at the same interviews and sees something completely different, that’s worth knowing before you publish anything.
Be clear about which categories mean what. A lot of disagreement between researchers comes down to fuzzy definitions, not real differences of opinion.
Why this matters outside of universities
This isn’t just an academic detail. NGOs use this kind of research to understand communities. Businesses use it to understand customers. Government projects use it to understand what’s actually working on the ground. If the research only reached the first, easier kind of “enough,” the conclusions built on top of it can end up sounding more certain than they really are, and decisions get made on shakier ground than anyone realizes.
Learning to do this properly
Knowing how to click through software like NVivo or ATLAS.ti is one thing. Knowing when you’ve truly understood your data, not just heard it, is a different skill, and it’s the one that actually makes research trustworthy.
Our NVivo, ATLAS.ti, QDA Miner & Report Writing course teaches exactly this — not just the software, but how to know when your analysis is actually solid enough to stand behind.
Want to move past just hearing what people say and actually understand what they mean? Explore our NVivo, ATLAS.ti, QDA Miner & Report Writing course and learn to do research people can actually trust.