{"id":"dat-pm-churn-reasons-tally","audience":"product-managers","category":"data","technique":"context","title":"Sort churn survey answers into reason categories","free":false,"task":"Sort the 12 churn survey answers into Price, Missing feature, Competitor and Other. Your team's rules: if they name another product, it's Competitor; if they give two reasons, use the first one. Write exactly 4 numbered lines, each as Category: count (#response numbers), then one unnumbered last line: Total: 12. Plain text, under 60 words.","materialLabel":"Churn survey export (attached to your prompt automatically)","material":"CANCELLATION SURVEY - \"Why did you cancel?\" (October)\n\n#1 too expensive for our team size\n#2 we moved to Trellis\n#3 no Gantt charts\n#4 budget cuts, can't justify the cost\n#5 Trellis had the integrations we needed\n#6 company shut down\n#7 needed offline mode for field crews\n#8 price went up 30% at renewal\n#9 switched to Planwise\n#10 missing time tracking\n#11 just didn't use it enough\n#12 too pricey, also missing recurring tasks\n\nNotes:\n- 48 accounts canceled in October; 12 answered.\n- Sales thinks \"it's all price.\" Let's see.\n- Roadmap review is Nov 18.","rules":[{"type":"numberedItems","count":4},{"type":"plainText"},{"type":"maxWords","max":60}],"mistakes":[{"title":"Not giving your rules","why":"The AI files #5 under missing features and double-counts #12.","fix":"Add: \"Naming another product = Competitor. Two reasons = use the first.\"","checks":["r0","r1"]},{"title":"Not asking to see which is which","why":"Bare counts can't be checked.","fix":"Add a line to copy: \"1. Price: 4 (#1, #4, #8, #12)\"","checks":["numberedItems","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 line: \"each line under 10 words.\"","checks":["maxWords"]},{"title":"Forgetting where it goes","why":"AI loves markdown tables and bold, which show up as clutter in a roadmap doc.","fix":"Add: \"Plain text, no bold, headings or tables.\"","checks":["plainText"]}],"promptHints":[{"label":"Gives the categories","patterns":["price","missing feature","competitor","other"]},{"label":"Gives the tie rules","patterns":["first (one|reason)","names? (another|a) product|competitor"]},{"label":"Asks for response numbers","patterns":["#|response numbers","which responses"]},{"label":"Sets a word limit","patterns":["\\d+\\s*words","word (limit|count)","under \\d+"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no (headings|tables)"]}],"hint":"Some answers fit two categories. Tell the AI your rules for those, and ask it to list response numbers so you can check.","rubricLabels":["#5 is counted as Competitor (it names Trellis) and #12 as Price (its first reason)","Counts are Price 4 (#1, #4, #8, #12), Missing feature 3 (#3, #7, #10), Competitor 3 (#2, #5, #9), Other 2 (#6, #11)","Each line lists the response numbers in parentheses, and nothing comes before the first numbered line"]}