{"id":"cs-training-survey-dedupe-average-percent","audience":"customer-success","category":"support-work","technique":"verification","track":"support","level":"intermediate","title":"Compute a training survey average after removing a duplicate","free":false,"task":"Write three lines starting \"Responses counted:\", \"Average:\" and \"Satisfied (4-5):\". Count each respondent once (a duplicate submission counts once). Average is to one decimal. Satisfied shows the count, the total and the whole-number percent, like \"N of M (P%)\". Plain text, under 40 words.","materialLabel":"Your material (attached to your prompt automatically)","material":"TRAINING SURVEY - \"How useful was the training?\" (1 = not useful, 5 = very useful)\nSession Nov 4, Halvorsen Co. Submissions as exported:\nr1: 5\nr2: 4\nr3: 3\nr4: 5\nr5: 2\nr6: 4\nr7: 4\nr7: 4 (second submission by the same person, 2 minutes later)\nr8: 5\nr9: 3\nr10: 4\n\nNote: 11 rows are in the export, 10 people were invited, and 10 attended.\nHelper: counting each of r1 to r10 once, the scores add up to 5+4+3+5+2+4+4+5+3+4 = 39 across 10 people.","rules":[{"type":"plainText"},{"type":"maxWords","max":40}],"mistakes":[{"title":"Counting the duplicate","why":"The AI uses all rows in the export.","fix":"Say: \"Count each respondent once; remove the duplicate.\"","checks":["r1","r2"]},{"title":"Wrong total","why":"The AI divides 35 by 10 or 39 by 9.","fix":"Say: \"Divide the total by the number of people counted.\"","checks":["r1"]},{"title":"Skipping the format","why":"The AI writes a paragraph.","fix":"Say: \"Three lines starting with these labels.\"","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 three labels","patterns":["responses counted","average","satisfied"]},{"label":"Removes duplicates","patterns":["duplicate","once"]},{"label":"Format for percent","patterns":["whole.number","67%","of 9"]},{"label":"One decimal","patterns":["one decimal"]},{"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":"Survey math starts with clean counts. Tell the AI to remove duplicates and give the output format.","mission":"Report the training survey results with the duplicate removed.","musts":["Count each respondent once","Average to one decimal","Satisfied as N of M"],"rubricLabels":["The output has exactly three lines starting \"Responses counted:\", \"Average:\" and \"Satisfied (4-5):\"","Responses counted is 10 (the duplicate r7 is counted once) and Average is 3.9","Satisfied (4-5) is 7 of 10 (70%)"]}