{"id":"mkt2-checkout-test-revenue-per-visitor","audience":"product-managers","category":"data","track":"marketing","technique":"verification","level":"advanced","title":"Pick the winning checkout variant by revenue per visitor, with math","free":false,"task":"Your checkout page test has three variants and the goal metric is revenue per visitor, not conversion rate. Write a prompt for a markdown table with columns Variant | Orders | Conv. rate | Rev/visitor, then one line starting \"Winner:\" that names the variant with the highest revenue per visitor and a reason under 15 words. Show conversion to one decimal and revenue per visitor in dollars and cents. No other text, under 90 words.","materialLabel":"Test results (attached to your prompt automatically)","material":"CHECKOUT TEST - three layouts, 2 weeks\nVisitors per variant: 5,000 (traffic split evenly)\nGoal metric chosen before the test: revenue per visitor (total revenue / visitors)\n\nVariant A (control, current single page)\n  Orders: 150   Revenue: $7,500\nVariant B (express wallet button first)\n  Orders: 165   Revenue: $7,400\nVariant C (shipping options hidden until step 2)\n  Orders: 140   Revenue: $8,400\n\nNotes:\n- Conversion rate = orders / visitors.\n- B has the most orders; the product team prefers B because it's the \"simplest\".\n- C's average order was higher: customers who bought added accessories.\n- Refund rate not yet available for any variant.\n- 'Statistical significance' has not been calculated.","rules":[{"type":"plainText"},{"type":"maxWords","max":90}],"mistakes":[{"title":"Choosing the most orders","why":"The AI crowns B since it converts best.","fix":"Add: \"The winner is the highest revenue per visitor, not conversions.\"","checks":["r1"]},{"title":"Skipping the formula","why":"Without the formula the AI rounds loosely.","fix":"Say: \"Rev/visitor = revenue / 5,000; conversion = orders / 5,000.\"","checks":["r0"]},{"title":"Claiming significance","why":"The AI says 'statistically significant'.","fix":"Say: \"Don't claim significance; it hasn't been calculated.\"","checks":["r2"]},{"title":"Letting markdown through","why":"AI adds bold text and headings by default.","fix":"Add: \"Plain text only, no bold, no headings.\"","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and often runs past a word limit.","fix":"Cap each part (for example \"each line under 15 words\") so the total lands well under the limit.","checks":["maxWords"]}],"promptHints":[{"label":"Names the table columns","patterns":["variant.*orders.*conv.*rev"]},{"label":"Defines the goal metric","patterns":["revenue per visitor","rev/visitor","not conversion"]},{"label":"Gives the formulas","patterns":["revenue (divided by|/) (5,000|visitors)","orders (divided by|/)"]},{"label":"Fixes decimals","patterns":["one decimal","dollars and cents"]},{"label":"Starts a Winner line","patterns":["winner:"]}],"hint":"The goal metric decides the winner, not the biggest conversion rate. State the metric, formulas and format, and ban significance claims.","mission":"Pick the checkout variant that earns the most per visitor.","musts":["Judge by revenue per visitor","Conversion to one decimal","Dollars and cents"],"rubricLabels":["The table shows A 150 orders, 3.0% and $1.50; B 165 orders, 3.3% and $1.48; C 140 orders, 2.8% and $1.68","The Winner line names C, the variant with the highest revenue per visitor, with a reason under 15 words","The output has no text outside the table and Winner line and does not claim significance or call B the winner"]}