{"id":"cx-csat-themes-with-counts","audience":"customer-support","category":"support-work","technique":"verification","track":"support","level":"intermediate","title":"Summarize CSAT themes with exact counts and shares","free":false,"task":"Summarize the 20 tagged comments for your director, Mei. Write exactly 4 numbered lines, one per theme, from most to least common, formatted like \"1. Speed: 8 of 20 (40%)\". After them, one last line starting \"Watch:\" naming the theme whose count doubled from last month, with both counts. Plain text, nothing else.","materialLabel":"Your material (attached to your prompt automatically)","material":"CSAT COMMENTS, OCTOBER (theme tag assigned by QA)\nC01 Speed\nC02 Friendliness\nC03 Speed\nC04 Billing\nC05 App bug\nC06 Speed\nC07 Friendliness\nC08 Billing\nC09 Speed\nC10 Friendliness\nC11 Speed\nC12 App bug\nC13 Friendliness\nC14 Billing\nC15 Speed\nC16 Speed\nC17 Billing\nC18 Friendliness\nC19 App bug\nC20 Speed\n\nSEPTEMBER COUNTS (20 comments): Friendliness 5, Billing 4, Speed 4, App bug 7\n\nQA note: C09 mentions both speed and billing; QA tagged it Speed on purpose.\n\nSample comments for context: C01 'Replied in 10 minutes, wow.' C04 'I was billed after cancelling.' C05 'The mobile app crashed when I attached a file.' C13 'Sara was so kind and patient with me.' C20 'Fixed in one reply, thank you!'\n\nOctober ticket volume was 1,240; only 20 comments were sampled and tagged by QA this month. September also sampled 20.","rules":[{"type":"plainText"},{"type":"numberedItems","count":4}],"mistakes":[{"title":"Miscounting from the list","why":"The AI estimates counts from a long list instead of counting each tag.","fix":"Say: \"Count every tag. Use the tags as given. Show 'n of 20'.\"","checks":["r0"]},{"title":"Watching the wrong theme","why":"The AI flags App bug because it had the biggest September number.","fix":"Define it: \"Watch = the theme whose count doubled. Give both counts.\"","checks":["r1"]},{"title":"Adding commentary","why":"The AI adds a paragraph of insights after the list.","fix":"Say: \"Output only the 4 lines and the Watch line.\"","checks":["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":"Leaving the list shape open","why":"The AI picks its own number of items and its own layout.","fix":"Show the shape: \"exactly N numbered lines like: 1. Label: detail\".","checks":["numberedItems"]}],"promptHints":[{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Fixes the number of lines","patterns":["\\b\\d+\\b.{0,15}(lines|items|numbered)","numbered","exactly \\d+"]},{"label":"Shows the line format","patterns":["of 20","\\d+\\.","format"]},{"label":"Defines Watch","patterns":["watch","doubled"]}],"hint":"Counting tasks need a fixed line format and a clear definition for any flag. Show both.","rubricLabels":["The four numbered lines are Speed: 8 of 20 (40%), Friendliness: 5 of 20 (25%), Billing: 4 of 20 (20%) and App bug: 3 of 20 (15%), in that order","The last line starts with \"Watch:\" and names Speed with both counts, 4 in September and 8 in October","The output contains no text other than the four numbered lines and the Watch line"]}