TranskribatorTranskribator
Writing a problem statement from interviews without overgeneralizing
Product research

Writing a problem statement from interviews without overgeneralizing

Write a problem statement from user interviews that identifies the person, context, goal, obstacle, impact and evidence without assuming a solution.

The Transkribator team5-minute read
SOURCEA PROBLEM YOU CAN TESTWho: project coordinatorWhen: shares a summarywith a contractorObstacle: no accessto the projectConsequence: risk of sendingan outdated versionFROM WORDS TO RESULTS
Working with a real recordingIllustrated training exampleTemplate to copy
Product research

A problem statement should guide research

“We need easier exporting” already includes a judgment and a solution. The team could start improving a button while the user actually needs to review the result with someone who has no access. A useful statement keeps attention on the situation and the observable obstacle.

Interviews give you examples, and the analyst has to define what they support. One memorable story gives you a hypothesis. It doesn’t justify “always.” Link to the episodes and distinguish paraphrases from direct quotes.

Workflow

Write the statement in five parts

  1. Name a specific group

    Use a role or behavior that matters to the situation. Demographics are useful only when they change the task.

  2. Describe when it happens

    Name the event, setting and people involved. Examples include sharing results after a call or checking a report before sending it. This context makes it possible to test the statement.

  3. Record the goal and obstacle

    What is the person trying to do, and what can you observe getting in their way? Describe the obstacle without equating it with the absence of your proposed feature.

  4. Add consequences and evidence

    Describe the observed impact: an extra step, a delay, a mistake or giving up on the task. Link to timestamps and say what remains unknown.

Worked example

From general frustration to a usable problem statement

This is a fictional training example. The names, circumstances and outcomes are included solely to explain the method.

Excerpt from a training recording · 12:40
After the meeting I send a summary to a contractor. They can’t see the project, so I copy the table into an email. Sometimes an old version goes out.

Weak wording: “Users need easier exporting.” It provides no context, goal or consequence and jumps to a solution.

Working hypothesis: “After meetings, coordinators share a summary with an external contractor who has no access to the project. Manual copying risks sending an old version.” Describe the limits of the sample separately. The example is fictional.

Who: project coordinatorWhen: shares the summary with a contractorObstacle: no access to the projectConsequence: risk of an outdated version
A problem statement connects a person, a situation, an obstacle and an observable consequence.
ExcerptWorking noteChecks
WhoCoordinators in the interviews reviewedHow was the group formed?
ContextSharing the summary with an external contractorDoes the situation recur?
ObstacleNo access to the projectIs that a cause or a symptom?
ConsequenceManual copying and version mix-upsIs there an observed example?
Ready-to-use template

Problem statement template

Fill in the fields before discussing solutions.

Template for your workSelect and copy
User group:
Context / event:
User’s goal:
Observable obstacle:
Consequences:
Example of behavior:
Quote and timestamp:
Number of supporting interviews, counted using one rule:
Who is outside the scope:
What remains unknown:
One-paragraph problem statement:
What will count as confirmation in the next check:
Check the result

Check the scope of the statement

  • The wording doesn’t hide a proposed solution.
  • The group and context are specific.
  • An example supports the consequence.
  • Any generalization fits the observations available.
InterviewObservable episodeCodingWho, when and whyStatementObstacle and impactScopeHypothesis to test
The problem statement remains a hypothesis until the team checks it beyond the original examples.
Frequently asked questions

A closer look

Can the statement mention how the user felt?

Yes, as part of the context or impact if they clearly said it. Avoid amplifying the emotion with a label they didn’t use.

Do we need numbers?

Only when you have verified them. Include time and frequency if measured or clearly stated. Don’t turn someone’s rough estimate into an exact metric.

Keep working with the recording