Convert vague feedback into testable problem statements

Translate fuzzy stakeholder complaints into precise, observable root problems with clear metrics and validation tests.

You getExtracted Core Symptoms, unverified Assumptions and testable Problem Statement

Prompt

Prompt
Act as a senior product discovery and systems analyst. Transform the vague stakeholder complaints provided below into structured, testable problem statements.
- Separate observable symptoms from emotional venting, assumed solutions, and hearsay.
- Identify the affected user group and the specific step in their workflow that breaks.
- Formulate a falsifiable problem statement stating who, what barrier occurs, and the measurable business impact.
- Define one rapid test to prove or disprove the problem before committing engineering or operational resources.
Output format:
- Extracted Core Symptoms (bullet points)
- Unverified Assumptions (bullet points)
- Testable Problem Statement (single structured paragraph)
- Validation Experiment (hypothesis, metric, pass/fail threshold)
My stakeholder notes are below:

Example output

Example for: customer support reps complaining that the ticketing tool is slow and broken

Extracted Core Symptoms

  • Support agents report taking longer to resolve standard billing tickets during morning shift peaks.
  • Multiple browser tabs freeze when opening customer account history profiles.

Unverified Assumptions

  • The ticketing software itself needs to be replaced immediately.
  • The problem affects all customer issue types equally.
  • Server latency is the primary cause rather than browser memory usage or API timeouts.

Testable Problem Statement
Tier-1 support agents spend an average of 4.2 minutes waiting on customer profile data to render during peak hours (09:00-11:00 EST), which inflates overall handling time and causes backlog accumulation, because the account history widget runs unindexed database queries across closed transactions.

Validation Experiment

  • Hypothesis: Caching the last 30 days of billing history will reduce page render time by at least 60%.
  • Metric: Average page load time for customer profiles between 09:00 and 11:00 EST.
  • Pass/Fail Threshold: Pass if average load time drops below 1.5 seconds across 100 test lookups; fail if load times remain above 3.0 seconds.
Written by AI from this prompt. Your results will vary.

Paste raw survey responses, meeting transcripts, or stakeholder emails at the bottom of the prompt in Claude, ChatGPT, or Gemini. Ask the model to suggest three discovery interview questions to run with users as a follow-up.

Make it repeatable

Keep using this prompt without pasting it in every time.

Keep exploring

More in Critical Thinking & Decisions
Critical Thinking & Decisions

Draft a calm devil's advocate brief

Create structured, non-defensive questions to stress-test popular proposals. Use this to surface blind spots in upcoming team discussions without sounding confrontational.