{"id":"dat-mkt-nps-without-staff","audience":"marketing","category":"data","technique":"context","title":"Calculate NPS after removing staff test responses","free":false,"task":"You're reporting the customer survey's NPS. 12 of the score-10 responses came from staff testing the survey (not marked in the export), so remove them first. Promoters score 9-10, detractors 0-6, and NPS = % promoters − % detractors. Write exactly 3 numbered lines (promoters, detractors, NPS), each with the math. Plain text, under 50 words.","materialLabel":"Survey export (attached to your prompt automatically)","material":"CUSTOMER SURVEY - \"How likely are you to recommend Pinecrest Bank's app?\" (Oct)\n\nScore | Responses\n10 | 37\n9 | 20\n8 | 18\n7 | 12\n6 | 10\n5 | 6\n4 | 4\n3 | 2\n2 | 1\n1 | 1\n0 | 1\nTotal | 112\n\nNotes:\n- Survey went to app users who logged in at least once in October.\n- Top complaint in comments: login takes too long.\n- The vendor dashboard shows \"NPS: 29\" (calculated on all 112).\n- Staff were asked to test the survey link on launch day; their answers went into the live results.\n- Survey closed Oct 31 with 112 total responses.\n- A score of 7 or 8 counts as passive.\n- Last quarter's NPS, calculated the same way, was 18.\n- Results go to the product and support leads next week.","rules":[{"type":"numberedItems","count":3},{"type":"plainText"},{"type":"maxWords","max":50}],"mistakes":[{"title":"Not mentioning the staff tests","why":"The export doesn't mark them, so the AI's NPS is inflated.","fix":"Add: \"Remove 12 staff responses from the 10s first.\"","checks":["r0","r2"]},{"title":"Not defining the groups","why":"The AI may count 7-8 as promoters or 6 as passive.","fix":"Add: \"Promoters 9-10, detractors 0-6.\"","checks":["r1"]},{"title":"Not asking to see the work","why":"A bare NPS can't be checked against the dashboard.","fix":"Add a line to copy: \"1. Promoters: (25 + 20) ÷ 100 = 45%\"","checks":["numberedItems","r1"]},{"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 14 words.\"","checks":["maxWords"]},{"title":"Forgetting where it goes","why":"AI loves markdown tables and bold, which show up as clutter in a marketing update.","fix":"Add: \"Plain text, no bold, headings or tables.\"","checks":["plainText"]}],"promptHints":[{"label":"Removes the staff tests","patterns":["staff","12","test"]},{"label":"Defines the groups","patterns":["9-10|9 or 10","0-6|0 to 6"]},{"label":"Asks to show the math","patterns":["show (the |your )?(math|work)","x \\$|\\bx\\b.{0,10}=","calculation","step"]},{"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":"The export includes responses that aren't customers. Tell the AI how many to remove and how NPS groups work.","rubricLabels":["Removes the 12 staff responses so the total is 100 and score 10 has 25","Promoters are (25 + 20) ÷ 100 = 45% and detractors are (10 + 6 + 4 + 2 + 1 + 1 + 1) = 25 ÷ 100 = 25%","NPS is 45 − 25 = 20 (not the dashboard's 29), with nothing before the first numbered line"]}