Customer interview guide and evidence log

Use one brief for one marketing decision and one record per conversation. Store identifying information and raw notes in an appropriately restricted workspace. The worked example is fictional.

Research brief

• Decision owner and date:
• Decision to inform:
• Current explanation:
• Plausible alternative explanations:
• Relevant workflow or buying episode:
• Participant criteria:
• Outcomes or groups to include:
• Recruitment route and its limitations:
• Intended uses of the notes:
• Recording agreement, if applicable:
• Review condition:

Discussion guide

1. What was happening when this became something you needed to address?
2. Walk me through the last time you did this work.
3. What did you try first, and what happened?
4. What alternatives did you consider, including doing nothing?
5. Who participated in the decision and the work?
6. What made a step easy or difficult?
7. What changed after you tried the solution?
8. What remains unresolved?

Ask follow-ups using the participant's language. Avoid suggesting an answer. If the conversation becomes support or a demo, mark where the context changed.

Interview record

• Participant reference and role:
• Date and relevant episode:
• Trigger:
• Workflow sequence:
• Current or previous alternatives:
• Decision participants:
• Obstacle:
• Outcome:
• Uncertainty or recall limitation:
• Agreed next step:

Mark quotes distinctly. Do not present an approximate note as verbatim language. Interview participation does not establish permission to publish a name or quotation.

Fictional evidence record

Observation: an engineer described checking logs in two systems before deciding whether to rerun an API test. Interpretation: collecting context may create friction. Alternative interpretation: the second check provides a necessary safeguard. Follow-up: ask what each system contributes and which step, if any, the engineer would remove. No product benefit is established by this observation alone.

Synthesis and decision

• What we observed:
• What it suggests:
• Sample and recruitment limits:
• Decision or next bounded test:
• What would change our view:
• Owner and review condition:

Repeated observations in a selected group can support a hypothesis without establishing its prevalence across the market.

Question quality and closing check

• Summary offered back to the participant:
• Correction or missing context they supplied:
• Evidence record updated before synthesis:

Record the correction even if it weakens the original hypothesis.