{"id":"fpa-collections-forecast-by-bucket","audience":"fpa","category":"finance-work","track":"finance","technique":"verification","level":"intermediate","title":"Forecast collections from the AR aging using bucket collection rates","free":false,"task":"You're forecasting next month's cash collections from the AR aging. Write a prompt for plain text with lines starting \"Current:\", \"1-30:\", \"31-60:\", \"61-90:\", \"90+:\", \"Total expected:\" and \"Not expected:\", each with the math in $K. Use the historical collection rate for each bucket. Don't mention the collections team's optimism. Under 90 words. No bullets or numbering.","materialLabel":"AR aging and historical collection rates (attached to your prompt automatically)","material":"COLLECTIONS FORECAST INPUTS - November 2026, Kestrel Home Goods ($ thousands)\n\nBucket | AR balance | Historical collection rate within the month\nCurrent | 120 | 98%\n1-30 days past due | 40 | 90%\n31-60 | 20 | 70%\n61-90 | 8 | 40%\n90+ | 5 | 10%\nTotal AR: 193\n\nMethod: expected collections = balance x historical rate for each bucket. Not expected = total AR - total expected.\n\nNotes: the collections team believes 'we will get 95% of everything this month'. That is a target and not based on the history. Use the rates above only. Round to one decimal in $K.","rules":[{"type":"plainText"},{"type":"maxWords","max":90},{"type":"maxBullets","max":0}],"promptHints":[{"label":"Names the lines","patterns":["current:","1-30:","total expected","not expected"]},{"label":"Uses bucket rates","patterns":["98%","historical"]},{"label":"Asks for the total","patterns":["total expected"]},{"label":"Bans the belief","patterns":["(do not|don'?t).*(95%|team)"]},{"label":"Caps length","patterns":["\\d+\\s*words","plain"]}],"mistakes":[{"title":"Applying one rate to all","why":"The AI uses 95% on $193K and gets $183.4K.","fix":"Say: \"Use each bucket's historical rate only.\"","checks":["r0","r1"]},{"title":"Echoing the team's 95%","why":"The AI mentions the target.","fix":"Add: \"Do not mention the collections team's 95% belief.\"","checks":["r2"]},{"title":"No not-expected line","why":"The AI skips what isn't expected.","fix":"Add: \"Not expected: total AR minus total expected.\"","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","maxBullets"]},{"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 line short.","checks":["maxWords"]}],"hint":"Collections forecasts apply each bucket's own history. Give the formula, the lines and the ban.","rubricLabels":["The bucket lines show 120 x 98% = 117.6, 40 x 90% = 36.0, 20 x 70% = 14.0, 8 x 40% = 3.2 and 5 x 10% = 0.5","The line starting \"Total expected:\" shows 117.6 + 36.0 + 14.0 + 3.2 + 0.5 = 171.3","The line starting \"Not expected:\" shows 193 - 171.3 = 21.7, and the output never mentions the team's 95% belief"]}