Research on a two-week clock
Writing team:
Ruth Alvarez Principal designer
Tomas Nyberg Engineering leadWhat you can honestly learn in ten working days, and what you should stop pretending to learn.
The two-week constraint is real
Most teams do not have a quarter for discovery. They have the time between one commitment and the next, and research either fits inside that or it does not happen.
Narrow the question until it fits
A fortnight cannot answer what users want. It can answer whether the people we spoke to can complete the one task we are about to rebuild, and where they stop.
- One decision the research is allowed to change.
- Five to seven participants, recruited from the existing user list.
- A written finding per session, same day, before the memory fades.
Analysis is not a separate phase
Waiting until the end to synthesise is what turns two weeks into five. Write the finding after each session; by the last one the pattern is already visible or it is genuinely not there.
Say what you did not learn
The credibility of a short study comes from its stated limits. Seven people cannot tell you about pricing. Say so in the report, and the parts that do hold get taken seriously.
Ship the next release with a team that has done this before
We join in-flight products and leave them faster to change than we found them. Research, design and front-end, on one clock.
View services
