There is a specific kind of failure familiar to anyone who has run research inside a company. The study is sound, the report is clear, everyone thanks you, and the roadmap does not change.
The findings were not wrong. They simply arrived as information when what was needed was an argument.
Agree the question with the people who will decide
Research commissioned by the design team and delivered to the leadership team has a structural problem: the audience never agreed the question was worth asking.
Before fieldwork, get the decision-makers to state what they currently believe and what would change their mind. It takes half an hour and it is the highest-leverage half hour in the project. It surfaces the real disagreement, and it makes the eventual finding hard to dismiss, because the criteria were theirs.
Ask what would change their mind before you run the study. If the answer is nothing, do not run the study.
Lead with the decision, not the method
Most research reports open with methodology, sample, and approach. By the time the recommendation appears on slide nineteen, the room has formed its own view.
Invert it. Open with the decision at stake, the recommendation, and the two or three pieces of evidence that most directly support it. Method goes in an appendix for the people who want it, and some will, which is fine. Rigour needs to be available, not performed.
Show behaviour, not preference
Stated preference is weak evidence and everyone senior knows it, which is why quoting what users said they wanted rarely moves anyone.
What does move people is watching someone fail at a task they were confident about, or seeing eight of ten participants miss the same control. Thirty seconds of session video does more than a page of synthesis, because it removes the interpreter. Nobody argues with what they just watched.
- Clips of the failure, not summaries of the failure.
- Counts before adjectives: six of nine, not most participants.
- Contradictory evidence included, with your reading of it.
- One clear recommendation per finding, not a menu of options.
Be honest about what you did not find
Overstating certainty is the fastest way to lose an experienced audience. If eight sessions cannot support a claim about the whole market, say so and make the smaller claim well.
Counter-intuitively, naming the limits increases influence. It signals that you are reporting rather than campaigning, which makes the claims you do make harder to wave away.
Arrive before the decision is made
Perfect research delivered after commitment is archaeology. A rough answer in week two beats a rigorous one in week nine, because the decision was taken in week three.
So we ask when the decision will actually be made and work backwards, accepting a smaller study to hit that date. The purpose is not to produce knowledge. It is to make the decision better than it would otherwise have been.
This is the thinking behind our research and validation work. Written by the UXperts team. If this is a problem you are living with, tell us about it.