+91 7023139797 | info@dynmarkit.com Serving global clients in USD and GBP
Research Report

Interviewing for signal, not comfort

The interview mistakes that quietly destroy research quality, and a short checklist we use to avoid them.

Most of the value of a good research programme is decided in the first fifteen minutes of each interview. If the interviewer is well prepared, the conversation will reach the question that matters. If the interviewer is not, the interview will be polite, pleasant, and of almost no analytic value.

The standard failure mode is that interviewers ask for opinions rather than for behaviour. Opinions are cheap. Behaviour is expensive to change and therefore almost always honest. A well run interview asks what the respondent actually did last time, not what they think they would do next time.

Six rules we follow

  • Prepare a written hypothesis. You are not asking questions to be surprised. You are asking questions to confirm or falsify a view. Write the view down first.
  • Ask about the last time, not the typical time. Memory about the typical case is fiction. Memory about the last case is still reasonably accurate.
  • Let silence do work. Most interviewers rescue their respondents too quickly. The useful answer often sits two seconds into a silence.
  • Ask for the walk through. If someone tells you they made a decision, ask them to walk you through how the decision actually got made, step by step, including who was in the room.
  • Write the answer down as given. Do not paraphrase in the notes. Paraphrase introduces the interviewer's assumptions, and the point of the exercise is to keep those out.
  • Debrief within an hour. The single largest source of research decay is the gap between the interview and the write up. Close it.

These six rules are boring. They are also the difference between research that changes a decision and research that decorates a slide deck.

What we do differently

In our work, interview guides are short. The first version of a guide is never the final version. By the fifth interview in any programme, the guide has usually been rewritten twice because the evidence has moved the question. That evolution is the signal that the research is doing real work.

If your research programme ends with the same questions it started with, you have gathered information but not learned anything. That is the fastest way to know the work was not worth the money.

Next step for the reader

Where this report connects to our practice pages

Readers who want to see how the firm turns this thinking into an engagement can read the market research practice page, which sets out how a research engagement is scoped, costed, and delivered in practice. A related report is How to read a market that no one else is looking at, because the sharpest research work is usually the work that asks a question the market has not yet asked itself.

Bring us a question that matters.

Share a short brief and we will come back within one business day with an initial view and the shape of a possible engagement.

Contact Our Team