{"id":"dat-mgr-pulse-survey-points","audience":"managers","category":"data","technique":"context","title":"Report pulse survey changes in points, not percent","free":false,"task":"Your director wants your team's pulse survey readout in exactly 3 numbered lines: 1) the biggest drop, 2) the biggest gain, 3) the response rate this quarter vs last. Changes must be in percentage points (78% to 62% is −16 points), never percent change. Plain text, under 60 words, nothing else.","materialLabel":"Survey results (attached to your prompt automatically)","material":"Q3 PULSE SURVEY - Claims Ops team (% agree)\n\nQuestion | Last quarter | This quarter\nI know what's expected of me | 85% | 88%\nI have the tools I need | 78% | 62%\nMy manager gives useful feedback | 70% | 74%\nI'd recommend working here | 80% | 71%\nMy workload is manageable | 46% | 55%\n\nResponses: last quarter 23 of 24; this quarter 18 of 24.\n\nComments (selected):\n- \"The new claims system is slow and crashes.\"\n- \"Thanks for cutting the Friday reports.\"\n- \"More 1:1 time would help.\"\n\nNote: HR's dashboard shows \"Tools: −20.5%\" because it uses percent change. Our director asked for points.\n\nFile notes:\n- Shared with the team in the usual folder.","rules":[{"type":"numberedItems","count":3},{"type":"plainText"},{"type":"maxWords","max":60}],"mistakes":[{"title":"Not defining the change","why":"The AI mixes percent change and points, or copies HR's −20.5%.","fix":"Add: \"Show changes in percentage points, like 78% to 62% = −16 points.\"","checks":["r0","r1"]},{"title":"Leaving the shape open","why":"The AI walks through all five questions and quotes every comment.","fix":"Say: \"Exactly 3 numbered lines: biggest drop, biggest gain, response rate.\"","checks":["numberedItems","r0","r1"]},{"title":"Forgetting the response rate","why":"A drop from 23 to 18 responses changes how much the results mean.","fix":"Add: \"Line 3: response rate now vs last quarter, as a percent.\"","checks":["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 14 words.\"","checks":["maxWords"]},{"title":"Forgetting where it goes","why":"AI loves markdown tables and bold, which show up as clutter in a director's Slack.","fix":"Add: \"Plain text, no bold, headings or tables.\"","checks":["plainText"]}],"promptHints":[{"label":"Asks for points","patterns":["points?","not percent"]},{"label":"Defines the 3 lines","patterns":["biggest drop","biggest gain|biggest (rise|increase|improvement)","response rate"]},{"label":"Says how many lines","patterns":["\\b(3|three)\\b.{0,20}lines","numbered"]},{"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":"\"Change\" can mean two different things. Tell the AI you want points, and spell out what each line covers.","rubricLabels":["Line 1 is the biggest drop: tools I need, 78% to 62%, −16 points (not −20.5%)","Line 2 is the biggest gain: workload is manageable, 46% to 55%, +9 points","Line 3 gives response rate 75% this quarter (18 ÷ 24) vs 96% last quarter (23 ÷ 24), and there is no intro or extra lines"]}