How many interviews do you really need?
Saturation, segments and scale: how to size a qualitative study for the question you are asking.
“How many interviews do we need?” is the first question in almost every study. The honest answer: it depends on what you want to know.
For discovery, look for saturation
When you explore needs and motivations, stop when new interviews stop bringing new themes. In a focused audience this often happens surprisingly early; broad audiences need more.
For comparison, plan per segment
If you want to compare groups, such as new and loyal customers or two markets, plan enough conversations in each group to see its themes on their own, not just in the total.
For measurement, add scale
When you want to know how widespread a view is, combine the depth of interviews with a larger sample. AI-moderated interviews make hundreds of conversations practical, so you get the why and the how many in one study.
Watch the quality, not just the count
A smaller number of rich, honest conversations beats a large number of rushed ones. Recruit carefully, keep interviews focused and reward people fairly for their time.
