Quant or qual? How to combine surveys and interviews
Numbers tell you what is happening and how much. Users' own words tell you why. How to sequence and combine them so research leads to a decision.
- Quantitative research tells you what moves and how much. Qualitative research tells you why.
- Explore first, then size: interviews to find the causes, a survey to weigh them.
- Or the other way: a number moves, interviews explain it.
- Put open and voice answers inside your surveys: you get the why at the scale of the what.
Two questions, two methods
A dashboard shows that conversion dropped. It cannot tell you why. Interviews tell you why, but not how many users share that reason. Each method answers a different question, which is why using only one leaves teams guessing.
Qual first, then quant: explore and size
When you do not know the causes yet, start with interviews. Then turn what you heard into a survey to measure each cause.
This is how the Betclic churn study worked: 12 interviews surfaced the causes, then 267 players surveyed sized them. Every cause was checked against two sources before it reached the roadmap.
Quant first, then qual: explain a number
When a metric moves and nobody agrees on why, start from the number: who is affected, since when. Then interview those users specifically. The quant tells you where to look, the qual tells you what you are looking at.
Both at once: open answers inside surveys
A survey does not have to be only checkboxes. One or two open questions, or voice answers, give you the reasons at scale.
In a two-wave study for a European betting and casino app, only 1% of players ranked personalised recommendations among their top criteria to pick a game, yet 85% wanted them in their ideal lobby. The numbers alone looked contradictory. The players' own words resolved it: they did not ask for “personalisation”, they asked to see games in their style and to save time.
Make it recurring
Running the same short survey every month turns research into a trend: you see what moved since last month and read why in the same place. That is the idea behind Monthly Pulse.
Questions teams ask.
- Should I start with a survey or with interviews?
- Start with interviews when you do not know the causes yet. Start with data or a survey when a number has moved and you need to know who is affected before asking why.
- Can AI replace interviews?
- AI helps to transcribe, tag and search what users said. It does not replace talking to the right users with the right questions, nor the judgment needed to turn findings into a decision.
- What is a good quant + qual setup for a small team?
- A short monthly survey with one or two open or voice questions, plus a handful of interviews when a number moves. It is cheap, continuous, and keeps the team close to users.