Score my forecast accuracy and find where budget owners are biased
A forecast accuracy scorecard by line item, with bias patterns, root causes, and process fixes.
The prompt
Review the accuracy of my past forecasts and help me improve the process.
Forecast history: [PASTE: forecast vs. actual by month/quarter for key lines - revenue, COGS, opex by department, cash]
Forecast horizon to assess: [e.g. "1-month-ahead and 3-month-ahead forecasts"]
How forecasts are built today: [WHO submits, METHOD - driver-based / run-rate / bottom-up]
Accuracy target, if any: [e.g. "within 3% on revenue"]
Analyze:
1. Error by line and period: absolute error, % error, and MAPE.
2. Bias: is each line consistently over- or under-forecast? Flag lines where most errors point the same way.
3. Whether accuracy gets worse at longer horizons, and by how much.
4. Likely root causes for the worst lines (sandbagging, timing slips, missing drivers, one-offs, late inputs).
Return:
- A scorecard table: Line | MAPE | Bias direction | Bias size | Trend | Likely cause.
- The 3 lines whose fixes would most improve overall accuracy.
- Specific process changes (driver changes, ownership, cut-off dates, ranges instead of point estimates).
- A simple monthly accuracy tracker layout I can maintain.
Do not guess causes as facts; label them as hypotheses to confirm with budget owners.
Highlighted [brackets] are blanks to replace with your own details.
Fill in the blanks and use this promptMore Finance & Accounting prompts
What to fill in
- e.g. "1-month-ahead and 3-month-ahead forecasts"
- WHO submits, METHOD - driver-based / run-rate / bottom-up
- e.g. "within 3% on revenue"
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
- Cut the number of days it takes to close the books each monthA diagnosis of the month-end close timeline with bottlenecks, quick wins, and a plan to close faster.
- Build a budget vs. actual variance analysis in ExcelSets up a budget vs. actual sheet with correct variance formulas, favorable/unfavorable logic, materiality flags, and a YTD view.
- Build a profit-and-loss estimate for a book acquisitionA title P&L that tests the advance, print run, pricing, and royalties to show the break-even point and whether an offer makes sense.
- Prepare a Good Faith Estimate for a self-pay patientBuilds a No Surprises Act Good Faith Estimate for uninsured or self-pay patients with required elements, timing, and a plain-language cover note.