Reporting findings honestly
🎯You'll be able to
Write a short research report that leads with what matters, shows evidence, recommends changes and states its limits.
🏠The everyday picture
A doctor's report says what was found, how sure they are, what to do next and what the tests couldn't rule out. A research report needs the same parts.
📘What it really is
- Lead with the summary: the two or three things that matter most, in plain words, with the key numbers.
- Task results as a small table of counts.
-
Findings ordered by severity, each with: what happened, why it matters, evidence (participant IDs and a quote) and a recommended change.
-
Recommend, don't design in the report. "Make the selected slot clearly different and announce it" leaves room for the design work (UX8).
-
Limits: who was and wasn't included, how many sessions, whether data is real or practice, and anything that could have affected results (a prototype that wasn't finished, a moderator who helped too soon).
-
Good news too: say what worked (P1 found the Sunday filter easily), so the team doesn't break it.
- Privacy: IDs, not names; blur or cut clips before sharing.
📝A worked example
Input: your counts (Lesson 3) and ratings (Lesson 2).
Steps:
- Write a three-sentence summary: only 2 of 5 booked without help; the slot picker and errors-by-colour were the biggest problems; the survey agrees that finding a free slot is the most common complaint.
- Fill the findings table in
research/findings.mdwith ID, tag, finding, evidence, severity and change. - Write the Limits section.
- Run
npm run test:research.
Expected output: 8 of 8 research tests pass. The reference solution has eight findings, two of them critical.
🌍Where it's used
- Product teams present a one-page summary at planning, with the full report linked.
- Clients of agencies receive a report and a short video of key moments (with consent and anonymised).
- Public bodies publish research summaries so that decisions can be checked.
📝Case scenario
A 40-page report on a new hospital kiosk sat unread. The researcher rewrote it as one page: three critical problems, the counts, a clip for each, and the recommended changes. Two of the three were fixed before launch.
🖱️Try it
Write the Summary and Limits sections of research/findings.md. Success: the summary fits in four sentences and includes counts; the
limits mention five participants, three recruiting places and practice data.
⚠️Common mistakes
- Burying the main point. Sign: method details on page one. Fix: summary first.
- No limits. Sign: findings presented as universal truths. Fix: say who was included and how many.
- Solutions disguised as findings. Sign: "Finding: we need a calendar". Fix: describe the problem; recommend a direction.
🛠️For your project
Project 1's test 8 checks for a Limits section that mentions the sample size and that the data is practice data.
✅Quick check
What comes first in a research report?
Show the answer
A short summary of what matters most, with key numbers.
Why include what worked well?
Show the answer
So the team keeps it when they change the design.