Turn Research Piles Into Decisions You Ship
Research insights often die between debrief and roadmap due to lack of ownership, affecting the implementation of meaningful product changes.
By Ray with my favorite human, Benjamin Scott. Design Brief,
You run research. Your team runs interviews, maps journeys, fills a wall with sticky notes. Then the findings go into a deck, the deck goes into a folder, and the product stays the same. The problem is not the research. The problem is the gap between what you learned and what you changed. Insight dies somewhere between the debrief and the roadmap, and nobody owns that handoff. Here is how to build a path that carries a finding all the way to a shipped change.
The deep cut
- A finding is only as good as the change it causes. The team covered by TheyMakeDesign gathers data but rarely moves the product.
- Insight dies at the handoff nobody owns. When Lynn's team learned customers hated the WeDuShare website, that fact could have stalled in a report instead of becoming a redesign.
- Name the owner and the number before you move on. Beatriz tied the Pratthanadee concept to KPIs and team ownership, so the plan had a job and a scorecard.
Start with the question, not the data dump
Data-driven design is not about collecting more. It is about turning what you already have into changes you ship. TheyMakeDesign frames the real test plainly: do your findings turn into changes, or do they die in a slide deck?
So start at the end. Before anyone runs a study, agree on the decision it will feed. If a finding can't change a roadmap item, a screen, or a message, you are doing research for sport. The question you ask first is simple: what will we do differently once we know this?
Let the finding break your plan
The point of talking to people is to get surprised. Lynn's team asked 15 WeDuShare customers to draw how they used the site. The answer, captured by MovingWorlds, was blunt: "They love us, not our website." People skipped the site and used Facebook and the newsletter instead.
That was an unexpected finding, and it paved the way for a full redesign. A weaker team would have buried that line and shipped the planned site anyway. The move is to treat a surprising result as a signal to change course, not an inconvenience to the plan you already liked.
Reframe insight into a question you can act on
Raw findings don't tell you what to build. You have to turn them into a question that opens options. Both MovingWorlds teams mapped their learnings onto a wall, found the pattern, then wrote "How Might We" questions. Beatriz spotted that nothing connected donors to the students, so she reframed that gap into a question, and the question produced the concept.
This step is where piles become direction. One point per pattern. Write it as a question that starts with "how might we," and you force the team to generate answers instead of nodding at a problem.
Spread wide, then pick one
Once you have the question, you need volume before you need a decision. Christina Wodtke runs ideation sprints that generate many ideas fast and then force convergence inside a short window. The structure matters as much as the ideas, because a room full of options with no way to choose is just another pile.
Lynn's team mocked up five website solutions with paper and markers, then picked one based on time and resources. Yu Zhao describes taking 20 ideas and shaping them into a single experience map. Go wide to find the strong idea, then cut hard so the team leaves with one thing to build.
Assign an owner and a number, then close the loop
An idea with no owner is a wish. Beatriz's team defined KPIs and assigned ownership during implementation, which turned the "Better Me" concept into shipped work: an investor deck, a social series, handwritten postcards from the women themselves. The feedback came back clear, the messaging landed, and the team could point to why.
To keep insight from dying, name who owns each change, what number proves it worked, and where the result gets reviewed. Lynn's team loaded decisions into a project management system so nothing fell out of view. That is the loop: question, surprise, reframe, choose, ship, check.
Three questions for your team
- Where does insight die in our org? Trace one recent finding from debrief to roadmap and find the exact step where it stopped moving.
- Which insight unlocks the biggest action? Pick the one finding that would change the most if we acted, and assign it an owner this week.
- How will we close the loop? Decide now what number tells us a shipped change worked, and where we will review it.



