{"id":"exec-two-source-conflict-flags-summary","audience":"consultants","category":"leadership-work","track":"leadership","level":"advanced","technique":"verification","title":"Compare two reports on the same metric and flag the conflicts","free":false,"task":"Two reports describe the same business. Compare them in a markdown table with columns Metric, Report A, Report B, Conflict? (Yes or No) and Which to trust. Include exactly the 5 metrics named in the notes. Which to trust must be A, B or Neither and use only the reasons in the notes. Then one line: Questions for the client: with a single question. Output only the table and that line.","materialLabel":"Two reports (attached to your prompt automatically)","material":"Client: Greenway Grocers (fictional)\n\nREPORT A (Finance, signed off, audited, Oct 31)\n- Revenue FY: $120.4M\n- Stores open: 82\n- Employees: 1,640\n- Gross margin: 24.1%\n- Customer count (loyalty members): 410,000\n\nREPORT B (Marketing deck, Nov 3, unaudited, rounded)\n- Revenue FY: $125M (includes a pending acquisition of 3 stores that closes Jan 15)\n- Stores open: 85 (includes the 3 pending stores)\n- Employees: 1,640\n- Gross margin: 24%\n- Loyalty members: 450,000 (includes lapsed members inactive for over 12 months)\n\nNotes: Marketing's numbers are rounded for presentation. Where they include pending or lapsed items, they do not describe today's business.","rules":[{"type":"maxWords","max":170}],"mistakes":[{"title":"Flagging rounding as conflict","why":"The AI marks 24.1% vs 24% as a conflict.","fix":"Say: \"Rounding differences are not conflicts.\"","checks":["r1"]},{"title":"Trusting the deck","why":"The AI averages the two.","fix":"Say: \"Trust the signed-off audited report where B includes pending or lapsed items.\"","checks":["r1"]},{"title":"Wrong columns","why":"The AI changes the headers.","fix":"Say: \"Use the five exact columns.\"","checks":["r0"]},{"title":"No question","why":"The AI omits the last line.","fix":"Say: \"End with Questions for the client: and one question.\"","checks":["r2"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting. Ask for a total word limit and it often lands over it.","fix":"Cap each cell: \"each cell under 14 words.\"","checks":["maxWords"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting. Ask for a total word limit and it often lands over it.","fix":"Cap each part instead, for example: \"each line under 12 words.\"","checks":["maxWords"]}],"promptHints":[{"label":"Columns","patterns":["metric","report a","report b","conflict","which to trust"]},{"label":"Five metrics","patterns":["5 metrics","five"]},{"label":"Rounding","patterns":["rounding","not a conflict"]},{"label":"Reasons","patterns":["only .* notes","reasons?"]},{"label":"Question line","patterns":["questions for the client"]}],"hint":"Name the columns, count the rows, and explain rounding and the final question line.","rubricLabels":["The table has columns Metric, Report A, Report B, Conflict? and Which to trust with exactly five rows for revenue, stores, employees, gross margin and loyalty members","Employees is No conflict, gross margin is No conflict (24.1% vs 24% is rounding), and revenue, stores and loyalty members are conflicts where A is trusted because B includes pending or lapsed items","The last line starts with \"Questions for the client:\" and asks one question, and no reason in the table goes beyond the notes"]}