{"id":"cs-qbr-exec-paragraph-verified-numbers-only","audience":"executives","category":"support-work","technique":"grounding","track":"support","level":"advanced","title":"Write an executive paragraph from verified numbers only","free":false,"task":"Write the QBR executive paragraph in exactly 2 lines. Line 1 starts \"Summary:\" and is one paragraph of at most 3 sentences using only customer-verified numbers. Line 2 starts \"Excluded:\" and lists, in a few words each, the unverified or estimated items you left out. Plain text, under 90 words.","materialLabel":"Your material (attached to your prompt automatically)","material":"QBR DATA, Calder Health (for the customer's executive team)\nCUSTOMER-VERIFIED: appointment no-shows fell from 14% to 9% (Calder's own scheduling report). Front-desk call volume fell from 1,200 to 900 calls a month (Calder's phone logs). 31 of 36 front-desk staff use the platform weekly (usage data, confirmed by Calder's admin).\nESTIMATES / UNVERIFIED: 'projected savings of $120,000 next year' (our model, not agreed by Calder); 'patient satisfaction up' (no survey data); 'industry-leading adoption' (marketing phrase).\nINTERNAL: renewal risk is low; upsell planned for Q2.\nThe executive team asked for numbers they can repeat to their board.","rules":[{"type":"plainText"},{"type":"maxWords","max":90}],"mistakes":[{"title":"Including the projection","why":"The $120,000 estimate sounds persuasive.","fix":"Say: \"Use only customer-verified numbers.\"","checks":["r1"]},{"title":"Missing the Excluded line","why":"The AI drops the unverified items silently.","fix":"Say: \"Line 2: Excluded: list what you left out.\"","checks":["r2"]},{"title":"Leaking internal notes","why":"The AI mentions renewal risk or the upsell.","fix":"Add: \"Nothing internal.\"","checks":["r0","r2"]},{"title":"Getting markdown back","why":"AI adds bold and headings by default, which show up as stray symbols where this is pasted.","fix":"Say \"plain text, no bold, no headings, no asterisks\".","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and usually runs long, so a vague \"keep it short\" lands well over the limit.","fix":"Cap each part instead of the total, for example \"each line under 15 words\".","checks":["maxWords"]}],"promptHints":[{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Sets a length limit","patterns":["\\d+\\s*words","word (limit|count)","under \\d+"]},{"label":"Names the two lines","patterns":["summary:","excluded:"]},{"label":"Limits to verified numbers","patterns":["verified","only"]}],"hint":"Executive paragraphs should be defensible. Limit to verified numbers, ask for an Excluded line, and keep internal items out.","rubricLabels":["The output is exactly two lines, the first starting Summary: with at most 3 sentences and the second starting Excluded:","Summary uses the verified numbers (no-shows 14% to 9%, calls 1,200 to 900 a month, 31 of 36 staff weekly) and no estimate","Excluded names the projected $120,000 savings, the satisfaction claim and the industry-leading claim, and the output mentions neither renewal risk nor the upsell"]}