Find recurring patterns across incidents, complaints, or tickets
Analyzes a pasted list of incidents, complaints, or tickets to find the patterns, clusters, and repeat causes behind them.
The prompt
Find the recurring patterns in this list of incidents.
The records: [PASTE THE INCIDENTS, COMPLAINTS, TICKETS, ERRORS, OR NEAR MISSES - ideally with date, location or team, category, description, and outcome]
Time period covered: [e.g. last 6 months]
What I want to understand: [e.g. "why the same problems keep coming back", "where to focus prevention"]
Anything that changed during the period: [NEW SYSTEM, STAFF CHANGES, POLICY, SEASONAL PEAKS]
Steps:
1. Group the records into themes based on what actually happened, not only on the category labels (labels are often inconsistent). Show how many records fall in each theme.
2. Look for patterns by time (day, shift, month), location or team, product or process step, customer type, and severity.
3. Spot repeats: the same issue at the same place, the same underlying cause showing up under different labels, and issues that got worse over time.
4. For the main patterns, suggest likely underlying causes as hypotheses to check, not conclusions.
5. Note data problems: missing fields, vague descriptions, inconsistent categories, and what to record in future to make analysis easier.
Return:
1. Summary: the 3 most important patterns in plain language.
2. A table: Theme | Count | Share of total | Example records | Trend | Possible cause.
3. Other patterns (time, place, team).
4. Where to focus first and why.
5. Recommended changes to how incidents are logged.
Rules: count only what's in the data and show your counts. Don't name or blame individuals; focus on processes and conditions. Remove personal details before pasting if the records contain them.
Highlighted [brackets] are blanks to replace with your own details.
Fill in the blanks and use this promptMore Data & Analysis prompts
What to fill in
- e.g. last 6 months
- NEW SYSTEM, STAFF CHANGES, POLICY, SEASONAL PEAKS
How this prompt is built
This prompt follows the Few-Shot structure (Few-shot prompting). A few-shot prompt includes one or more worked examples of the input and the output you want, so the model can copy the pattern instead of guessing it.
Related prompts
- Brainstorm and prioritize features for a predictive modelA ranked list of candidate features grounded in the problem, with leakage checks, build effort, and expected signal.
- Turn a business question into testable hypothesesA prioritized backlog of hypotheses with the data needed to test each.
- Design an A/B testHypothesis, variants, metrics, sample size, and duration for a valid test.
- Break down my results into the business drivers behind themBuilds a driver tree from your results, sizes how much each driver contributed to the change, and names which ones to act on.