An Opportunity Solution Tree from interviews: find opportunities before proposing features
Put one desired outcome at the top, then opportunities in users’ own words. Add several potential solutions for the selected opportunity and tests of their assumptions. This keeps the first idea from becoming the only option.
How to build an Opportunity Solution Tree from interviews
Put one desired outcome at the top, then opportunities in users’ own words. Add several potential solutions for the selected opportunity and tests of their assumptions. This keeps the first idea from becoming the only option.
A transcript is a verifiable source of wording. It preserves context, qualifiers and the order of the reasoning. The author still needs to decide on the final wording and test it with participants.
Build the tree step by step
Name the outcome
Write down one measurable outcome the team is accountable for.
Identify opportunities
Group needs, pain points and wants from the interviews. Phrase them in terms of what people experience.
Choose a branch
Use the available data to compare importance, frequency, strategic fit and uncertainty.
Consider different solutions
For the chosen opportunity, propose several options and write down the critical assumptions behind each one.
Example: getting to a useful part of a long recording faster
This is a fictional training example. The names, circumstances and outcomes are included solely to explain the method.
“I need one answer from an hour-long conversation. I don’t need the whole document.”
An opportunity is “quickly find an answer in a long recording.” “Add smart search” is already a solution. Give it a place under the need it addresses.
For the same opportunity, the team might consider text search, a table of contents with timecodes and questions marked in advance. Each idea has its own assumptions about query quality, conversation structure and the cost of manual review.
| Level | Check question | Example |
|---|---|---|
| Outcome | What change do we want to see? | Find the passage faster |
| User | What gets in the user’s way? | The answer gets lost in a long recording |
| Solution | How could we help? | Search, contents, tags |
| Test | What needs to be true? | Users know the words to search for |
Tree branch template
Copy the template and fill it in using a specific recording. Label guesses as hypotheses.
Desired outcome: Opportunity in the user’s words: Evidence and timecodes: Alternative solutions: Critical assumptions: Next test:
Four checks before sending
- There is one outcome at the top.
- Opportunities are described without feature names.
- There are several solutions for the selected opportunity.
- Each solution has a testable assumption.
A closer look
Can we build a tree from one interview?
You can start a draft. Mark where the evidence is thin and avoid treating an individual case as a market priority.
Where do feature requests go?
Find out what the person is trying to accomplish. The request itself can stay as a candidate solution.