{"id":"data2-dashboard-launch-readiness-severity-lines","audience":"managers","category":"data","track":"data","technique":"constraints","level":"intermediate","title":"Triage dashboard test findings into blockers and fix-later items","free":false,"task":"A new dashboard is meant to launch Monday and the tester found five issues. Write exactly 5 plain-text lines, one per finding, in the form '[Blocker] finding - why' or '[Fix soon] finding - why', each under 18 words. An issue is a Blocker only if numbers on the dashboard are wrong or someone who shouldn't have access does. Everything else is Fix soon.","materialLabel":"Tester findings (attached to your prompt automatically)","material":"Dashboard test log - Revenue dashboard, Friday Nov 7 (Stonebridge Furniture)\n\n1. The Revenue total is $1,200 higher than finance's ledger ($1,200,000 vs $1,198,800). Cause not yet known.\n2. The \"Region\" filter resets to All when the page refreshes.\n3. On a phone, the right-hand column of the table is cut off.\n4. Two metrics (Net Retention and Gross Margin) have no definition in the tooltip.\n5. The access list includes a former contractor who left in September; the dashboard shows customer names and revenue.\n\nNotes:\n- Launch is Monday morning; the dashboard goes to all managers.\n- The $1,200 gap is 0.1% of revenue.\n- The finance director said any difference from the ledger needs an explanation before launch.\n- Fixes for items 2, 3 and 4 are each estimated at under a day.","rules":[{"type":"plainText"},{"type":"maxWords","max":100}],"mistakes":[{"title":"Discounting the small gap","why":"The AI calls the $1,200 gap (0.1%) minor and tags it Fix soon.","fix":"Add: \"Any difference from the ledger counts as wrong numbers.\"","checks":["r0"]},{"title":"Ignoring the access issue","why":"The AI treats the former contractor as a cleanup item.","fix":"Add: \"Access for someone who shouldn't have it is a Blocker.\"","checks":["r1"]},{"title":"Tagging usability bugs as blockers","why":"The AI marks the phone layout and filter reset as Blockers.","fix":"Add: \"Everything else is Fix soon.\"","checks":["r2"]},{"title":"Leaving the format to chance","why":"The AI picks its own length and layout unless told the exact shape.","fix":"Add: \"Plain text, no bold or headings.\"","checks":["plainText"]},{"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 part instead, for example \"each line under 15 words.\"","checks":["maxWords"]}],"promptHints":[{"label":"Gives the two tags","patterns":["\\[blocker\\]","\\[fix soon\\]"]},{"label":"States the rule","patterns":["only if","wrong","access"]},{"label":"Says small gaps count","patterns":["any difference","counts"]},{"label":"Keeps the order","patterns":["same order","exactly 5"]},{"label":"Rules out extras","patterns":["no intro","no notes"]}],"hint":"A severity rule beats gut feel. Quote the rule and the line shape, and tell the AI that small differences still count.","mission":"Which dashboard issues should block launch?","musts":["Five tagged lines","Rule applied exactly","Order kept"],"rubricLabels":["Finding 1 (revenue $1,200 above the ledger) is tagged [Blocker] and finding 5 (former contractor still has access to customer names and revenue) is tagged [Blocker]","Findings 2 (Region filter resets), 3 (phone layout cut off) and 4 (missing tooltip definitions) are tagged [Fix soon]","There are exactly 5 lines in the form '[Blocker] finding - why' or '[Fix soon] finding - why', in the order of the findings"]}