Scattered paper scraps on the left, three tidy stacks in the middle and one folded card standing upright on the right.

2of 4

Synthesising one problem statement from customer research

Bring observations together around a particular person, need and reason. Keep your preferred solution out of the problem statement.

Build & Validate
Customer discoveryEvidence-led building

Adapted by the EWOR Team from an article on EWOR (opens in a new tab)4 min read

From customer notes to problem statement

Customer conversations leave you with fragments: quotes, observed actions, workarounds, exceptions and your own interpretations. Even a detailed set of notes can be misleading if it leaves room to see only what confirms your preferred idea.

Breaking notes into individual pieces is analysis. Bringing those pieces together into an explanation is synthesis. You need both: first understand the evidence, then turn it into a problem statement specific enough to direct ideas and tests. Keep in mind that the same evidence may support more than one interpretation.

Preserve the evidence before clustering it

Start by keeping three layers separate:

  • Reported: what the participant said.
  • Observed: what happened in their workflow or environment.
  • Inferred: what your team thinks the evidence might mean.

Keep the context and source attached to each note, and remove identifying details when they are not needed. Resist the urge to translate every observation into polished product language. The friction, sequence of events and customer's own words are often where the useful insights lie.

Before clustering notes into patterns, surface contradictions and edge cases. If a pattern only holds when you leave out inconvenient observations, it is worth a closer look.

Build patterns around situations and needs

  • User and context: who experiences the problem and in what situation.
  • Job: what they are trying to accomplish.
  • Trigger and sequence: what sets the problem off and what happens next.
  • Current workaround: how they deal with it today.
  • Consequence or unmet need: what the problem costs them or leaves unresolved.
  • Why the alternative persists: what keeps them using their current approach.

Avoid clustering around feature requests alone. "Users asked for a dashboard" tells you little about the underlying problem. A stronger pattern explains what they need to monitor, what decision they cannot make today and why their existing information falls short.

For each cluster, keep both supporting and conflicting evidence. A cluster built on vague opinions deserves less confidence than one backed by repeated behaviours in the exact context you are targeting.

  • User and context: who experiences the problem and in what situation.
  • Job: what they are trying to accomplish.
  • Trigger and sequence: what sets the problem off and what happens next.
  • Current workaround: how they deal with it today.
  • Consequence or unmet need: what the problem costs them or leaves unresolved.
  • Why the alternative persists: what keeps them using their current approach.

Avoid clustering around feature requests alone. "Users asked for a dashboard" tells you little about the underlying problem. A stronger pattern explains what they need to monitor, what decision they cannot make today and why their existing information falls short.

For each cluster, keep both supporting and conflicting evidence. A cluster built on vague opinions deserves less confidence than one backed by repeated behaviours in the exact context you are targeting.

Draft a point of view

A useful point of view tells you who the user is, what they need and why. A useful structure to follow is:

[Specific user in a specific situation] needs [outcome or capability] because [evidence-backed reason].

Keep your proposed product out of the problem statement. "Finance leads need a dashboard because dashboards are easier" jumps straight to a solution without explaining the underlying problem. A stronger statement identifies the decision or outcome the user struggles to achieve and what makes it difficult.

The same evidence can support more than one plausible problem statement, and that is fine. If your interpretation differs, write out the competing versions and identify what evidence would help you determine which one fits best.

Check the statement against your notes

A strong problem statement should pass five checks:

  1. Traceable: Can you link every part to an interview, observation or documented inference?
  2. Specific: Is the user and situation clear enough to recruit for a test?
  3. Problem-Led: Does it leave room for multiple solutions?
  4. Consequential: Does the evidence show why the problem matters?
  5. Revisable: Does it reflect your best current understanding, rather than claim to be the final truth?

If it falls short, go back to the evidence. You may need to narrow your target segment, reconsider the context or conduct more discovery. Better wording alone won't fix a weak foundation.

Hand over a question you can test

Finish the synthesis with a short evidence memo covering the problem statement, supporting patterns, counter-evidence, your level of confidence and the next assumption to test. This gives your team a clear handoff from discovery to experimentation.

A strong problem statement helps you decide what to test next without pretending discovery is complete. If new evidence challenges your understanding, update the statement rather than trying to explain the evidence away.

Source: EWOR article (opens in a new tab)

No comments yet.

Be the first to comment on this signal.

Get the next signal before everyone else.

Frameworks and contrarian theses for outlier founders, straight to your inbox. No fluff, no recycled advice, weekly.

The EWOR Signals newsletter

Continue learning

Related modules