{"id":"fpa-forecast-methods-gap-adjusted-memo","audience":"fpa","category":"finance-work","track":"finance","technique":"verification","level":"advanced","title":"Reconcile three revenue forecasts after known adjustments","free":false,"task":"You're preparing a memo for the CFO on three different full-year revenue forecasts. Write a prompt for plain text with headings on their own lines, in order: Raw, Adjusted, Spread, Open items. Apply only the adjustments stated in the material, show each adjusted figure, and do not pick a forecast or average them. Spread is the highest minus the lowest adjusted forecast, as a dollar amount and a percent of the lowest. Open items lists what to confirm. Under 150 words.","materialLabel":"Source material (attached to your prompt automatically)","material":"FULL-YEAR 2026 REVENUE FORECASTS ($ millions)\n\nMethod 1, run-rate (finance): 10.2\nMethod 2, pipeline-weighted (sales ops): 9.6\nMethod 3, top-down (strategy): 10.8\n\nKnown adjustments:\n- Method 1 includes a one-time $0.4M license sale in Q2 that will not repeat in the second half. Finance says the run-rate method annualized it by mistake; remove the 0.4.\n- Method 2 excludes subscription renewals of $0.5M that are certain to renew. Sales ops confirms the renewals are not in the pipeline report; add 0.5.\n- Method 3 has no known errors.\n\nThe board plan for the year is 10.0. The CFO wants to understand why the numbers differ before the Nov 24 forecast review.\n\nNote: strategy's top-down method assumes 8% growth over last year's 10.0 and 0.8 of that is not yet linked to any specific customer or product.","rules":[{"type":"plainText"},{"type":"maxWords","max":150}],"promptHints":[{"label":"Names the four headings","patterns":["raw","adjusted","spread","open items"]},{"label":"Asks for each adjustment","patterns":["adjust","show"]},{"label":"Bans picking or averaging","patterns":["do not pick","average"]},{"label":"Defines spread","patterns":["highest minus","lowest","percent"]},{"label":"Caps length","patterns":["\\d+\\s*words","under \\d+"]}],"mistakes":[{"title":"Averaging the forecasts","why":"The AI gives a blended number.","fix":"Add: \"Do not pick a forecast or average them.\"","checks":["r1"]},{"title":"Skipping an adjustment","why":"The AI removes the 0.4 but forgets to add the 0.5.","fix":"Add: \"Apply every stated adjustment and show each one.\"","checks":["r0"]},{"title":"Wrong spread base","why":"The AI uses the raw figures or divides by the highest.","fix":"Add: \"Spread = highest adjusted minus lowest adjusted, percent of the lowest.\"","checks":["r1"]},{"title":"No open items","why":"The AI ends with a conclusion.","fix":"Add: \"Open items: what to confirm.\"","checks":["r2"]},{"title":"Markdown extras","why":"The AI adds bold labels, headings and a closing note.","fix":"Say: \"Plain text, no bold. Output only the lines asked for.\"","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and runs past word limits.","fix":"Ask for fewer words than the real cap and keep each part short.","checks":["maxWords"]}],"hint":"Reconciling forecasts means applying only the stated fixes, then reporting the remaining gap, not choosing.","rubricLabels":["Raw lists 10.2, 9.6 and 10.8, and Adjusted shows Method 1 as 10.2 - 0.4 = 9.8, Method 2 as 9.6 + 0.5 = 10.1 and Method 3 unchanged at 10.8","Spread shows 10.8 - 9.8 = 1.0 ($1.0M), which is about 10.2% of the lowest (1.0 / 9.8), and the memo does not pick or average the forecasts","Open items names to confirm the unlinked 0.8 in Method 3 and the treatment of the one-time license sale and the renewals, as questions or confirmations"]}