{"id":"eval-mkt-sampling-plan-ai-product-descriptions","audience":"marketing","category":"evaluation","track":"evals","technique":"constraints","title":"Plan a weekly review of AI-written product descriptions","free":false,"task":"Your team can spend 3 hours a week reviewing AI-written product descriptions, at 2 minutes each. Write exactly 4 numbered lines: Capacity (the math), Always review (the high-risk groups from the data, with their total), Random sample (how many of the rest, drawn across every category), and Sample rate (the math as a percent). Plain text, under 100 words.","materialLabel":"This week's description numbers (attached to your prompt automatically)","material":"Larkspur Market - AI product descriptions, week of Nov 3\n\nTotal new or updated descriptions: 1,040\n\nHigh-risk groups (policy: review every one)\n- Brand-new products (never reviewed before): 25\n- Descriptions a shopper reported as wrong: 15\n\nEverything else: 1,000\n\nBy category (all 1,040)\nPantry 420 | Produce 180 | Household 240 | Baby 110 | Pet 90\n\nTeam notes\n- Last month Jen reviewed only Pantry, since it's the biggest.\n- A Baby product description listed the wrong age range in October.\n- Review form is pass/fail; each takes about 2 minutes.\n- The SEO team asked for keyword checks. Separate project.","rules":[{"type":"numberedItems","count":4},{"type":"plainText"},{"type":"maxWords","max":100}],"mistakes":[{"title":"Reviewing only the biggest category","why":"The AI accepts last month's habit of checking only Pantry.","fix":"Say: \"Random sample drawn across every category.\"","checks":["r2"]},{"title":"Forgetting the must-review items","why":"The AI samples at random and misses new products and shopper reports.","fix":"Add: \"Always review the high-risk groups first, then sample the rest.\"","checks":["r1"]},{"title":"Hiding the math","why":"The AI says 'about 90' or '5%' without showing how.","fix":"Say: \"Show the math on each line, like 3 x 60 / 2 = 90.\"","checks":["r0","r2"]},{"title":"A long review process","why":"The AI writes a QA workflow with roles and tools.","fix":"Be exact: \"Exactly 4 numbered lines: Capacity, Always review, Random sample, Sample rate.\"","checks":["numberedItems","maxWords","plainText"]}],"promptHints":[{"label":"Asks for the math","patterns":["math","show","x 60","calculat"]},{"label":"Reviews high-risk items first","patterns":["high.?risk","always review","new product","report"]},{"label":"Samples every category","patterns":["random","every category","all (five|5) categor","across"]},{"label":"Says exactly how many","patterns":["\\b(4|four)\\b.{0,25}(line)","exactly \\d+"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"A review plan has to fit your hours and still catch risky items. Start from capacity, cover the must-review groups, and sample the rest at random across every category.","rubricLabels":["Capacity shows 3 hours x 60 = 180 minutes / 2 = 90 descriptions","Always review names brand-new products (25) and shopper-reported descriptions (15) for a total of 40, and Random sample gives 90 - 40 = 50 from the other 1,000","The random sample is drawn across all five categories, not only Pantry, and Sample rate shows 50 / 1,000 = 5%"]}