Break down why a number changed between two periods
A driver breakdown and bridge that shows exactly which segments, volume, price, or mix effects caused a metric to move.
The prompt
Help me explain why a key number changed and which drivers caused it.
Metric: [e.g. revenue, gross margin, conversion rate, payroll cost, ticket volume]
Periods compared: [e.g. "Q3 vs. Q2", "this month vs. the same month last year"]
Data for both periods: [PASTE - broken down by whatever I have: product, region, channel, customer segment, price, volume]
Known events: [price changes, launches, outages, seasonality, one-offs, definition changes]
Audience: [WHO will hear the explanation]
Steps:
1. Calculate the total change in absolute and percentage terms.
2. Decompose it into drivers. For a rate or average, separate mix (the make-up of the base changed) from rate (performance within each segment changed). For revenue or cost, separate volume, price or rate, and mix. Make the parts add up exactly to the total.
3. Rank segments by their contribution to the total change, not by their own percentage change.
4. Separate one-off and timing effects from ongoing ones.
5. Flag any movement that could be a data or definition change rather than a real change.
Return:
1. A one-sentence answer: "[Metric] changed by [amount], mainly because ..."
2. A bridge table: Starting value | Each driver's contribution | Ending value, reconciling exactly. Describe how to show it as a waterfall chart in Excel or PowerPoint.
3. Top 5 contributors: Segment | Contribution | Explanation | Evidence | Confidence.
4. The unexplained remainder and what to check next.
5. A 3-bullet summary for my audience.
Use only my data. Where a driver is plausible but not shown in the data, label it [HYPOTHESIS - VERIFY] rather than stating it as the cause.
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. "Q3 vs. Q2", "this month vs. the same month last year"
- WHO will hear the explanation
- Metric
- amount
- HYPOTHESIS - VERIFY
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
- Analyze donor retention and find lapsed donors to win backA retention analysis using LYBUNT and SYBUNT segments, upgrade and downgrade patterns, and a prioritized win-back plan.
- Evaluate a statistical forecast's accuracy and explain the missesA backtest and error review of a demand or revenue forecast against simple benchmarks, showing where it misses and how to improve it.
- Find out why overtime is rising and how to cut itAn analysis of overtime data that separates demand, staffing, and process causes, with targeted actions to reduce it.
- Solve an Excel or Google Sheets problemGets you the right formula, pivot, or approach — with an explanation you can reuse.