Research without a stated decision expands to fill whatever time it is given. There is always another cohort to interview, another dataset to join, another edge case that might overturn the finding. Teams mistake this for rigour. It is more often avoidance, because a finding that has been written down can be proved wrong, and an ongoing study cannot be proved anything at all.
The fix is unglamorous. Before anyone books an interview we write one sentence: the decision this research exists to serve, and the person who will make it. Then every method question answers itself. Sample size is whatever would change that decision. Depth is whatever the decision maker needs in order to believe it. Anything that would not move the decision either way does not get funded.
The uncomfortable part is what this exposes. Perhaps a third of the time, writing the sentence reveals that the decision has already been taken and the research was commissioned to ratify it. That is worth discovering in week one rather than week nine, and it is a far better use of a research budget than a beautifully presented report that changes nothing.
Comments