{"id":"cs-qbr-slide-claims-vs-data-labels","audience":"customer-success","category":"support-work","technique":"verification","track":"support","level":"advanced","title":"Check five QBR slide claims against the data","free":false,"task":"Check each claim against the data table. Output five lines \"Claim 1:\" to \"Claim 5:\", each with a label and the number that supports it after a dash. Supported = the data meets or beats the claim. Overstated = the data moves the right way but by less than claimed. Not supported = the data contradicts the claim. Plain text, under 110 words.","materialLabel":"Your material (attached to your prompt automatically)","material":"QBR DATA - Q2 vs Q3 (Halvorsen Co)\nActive users: 380 -> 412\nSupport tickets: 52 -> 38\nAverage response time: 6.5 hours -> 5.0 hours\nNPS: 31 -> 38 (14 responses in Q3)\nUptime in Q3: 99.92%\n\nDRAFT SLIDE CLAIMS\nClaim 1: \"Active users grew by more than 10%.\"\nClaim 2: \"Tickets fell by more than a quarter.\"\nClaim 3: \"We cut response time in half.\"\nClaim 4: \"NPS improved by 7 points.\"\nClaim 5: \"We delivered 100% uptime in Q3.\"\n\nHelpers: 412 / 380 = 1.084. 38 / 52 = 0.731. 5.0 / 6.5 = 0.769.","rules":[{"type":"plainText"},{"type":"maxWords","max":110}],"mistakes":[{"title":"Rounding up the growth","why":"The AI says 8.4% is close enough to 10%.","fix":"Say: \"Overstated if the data moves the right way by less than claimed.\"","checks":["r1"]},{"title":"Calling 99.92% supported","why":"The AI treats near-100% as 100%.","fix":"Say: \"Not supported if the data contradicts the claim.\"","checks":["r2"]},{"title":"No numbers","why":"The AI gives labels only.","fix":"Say: \"Add the supporting number after a dash on each line.\"","checks":["r0"]},{"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":"The AI runs long and cannot count its own words.","fix":"Cap each part (\"each line under 12 words\") and add \"under N words in total\".","checks":["maxWords"]}],"promptHints":[{"label":"Names the labels","patterns":["supported","overstated","not supported"]},{"label":"Defines each label","patterns":["moves the right way","contradicts","meets or beats"]},{"label":"Asks for numbers","patterns":["number","after a dash"]},{"label":"Uses the helpers","patterns":["helper","8\\.4"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Sets a word cap","patterns":["\\d+\\s*words","under \\d+","word"]}],"hint":"Define each label precisely and require the number, so slide claims can be checked rather than accepted.","mission":"Check whether each of five QBR slide claims holds up.","musts":["Supported, Overstated or Not supported","Cite the supporting number","Under 110 words"],"rubricLabels":["The output has five lines Claim 1: to Claim 5: each with one of the labels Supported, Overstated or Not supported and a number after a dash","Claim 1 is Overstated (8.4% growth) and Claim 3 is Overstated (about 23% faster, not half)","Claim 2 is Supported (about 27% fewer tickets), Claim 4 is Supported (up 7 points), and Claim 5 is Not supported (99.92%, not 100%)"]}