{"id":"data2-dual-axis-chart-causal-claim-check","audience":"analysts","category":"data","track":"data","technique":"verification","level":"intermediate","title":"Check a \"ads drive sales\" claim against last year's numbers","free":false,"task":"A dual-axis chart slide claims 'Every extra $1K of ad spend brings $10K in sales'. Check it against both years' numbers. Write exactly 4 labeled lines in plain text, under 100 words: Claim, Checks out, Doesn't hold and Safer wording. Use only the figures given.","materialLabel":"Slide and monthly numbers (attached to your prompt automatically)","material":"Slide 9 - \"Ad spend drives sales\" (Maple & Pine Tea Co.)\n\nClaim on the slide: \"Every extra $1K of ad spend brings $10K in sales.\"\n\nThis year (Oct to Dec):\n- Ad spend: Oct $20K, Nov $30K, Dec $40K\n- Sales: Oct $200K, Nov $330K, Dec $480K\n\nSame months last year:\n- Ad spend: $25K every month (flat)\n- Sales: Oct $190K, Nov $310K, Dec $450K\n\nContext:\n- The chart uses two vertical axes, one for spend and one for sales, so the two lines look glued together.\n- The tea business sells most of its gift boxes in November and December.\n- No holiday promotions ran this year that did not also run last year.\n- The marketing director presents the slide Friday.","rules":[{"type":"plainText"},{"type":"maxWords","max":100}],"mistakes":[{"title":"Accepting the chart","why":"The AI agrees ads and sales move together, so ads drive sales.","fix":"Add: \"Use last year's numbers: flat spend, sales still up $260K.\"","checks":["r0"]},{"title":"Missing the seasonal point","why":"The AI doesn't say November and December are naturally busy.","fix":"Add: \"Say the rise is largely seasonal.\"","checks":["r1"]},{"title":"Weak rewrite","why":"The AI suggests 'ads help' as safer wording.","fix":"Add: \"Safer wording must not say ads caused the increase.\"","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 four labels","patterns":["claim:","checks out:","doesn.t hold:","safer wording:"]},{"label":"Uses last year","patterns":["last year","flat"]},{"label":"Names the seasonal point","patterns":["seasonal","season"]},{"label":"Gives the figures","patterns":["260K","280K","190K"]},{"label":"Limits the data","patterns":["only the figures","use only"]}],"hint":"Correlation claims fall apart against a baseline. Ask for last year's flat-spend comparison explicitly.","mission":"Is this ad spend claim right?","musts":["Four labeled lines","Use last year's baseline","Safer wording"],"rubricLabels":["Claim restates the $10K per $1K claim and Checks out says both spend and sales rose from October to December this year (spend $20K to $40K, sales $200K to $480K)","Doesn't hold uses last year's figures: with spend flat at $25K, sales still rose from $190K to $450K (+$260K), so the Oct to Dec rise is largely seasonal, and this year's rise (+$280K sales on +$20K spend) is not $10K per $1K","Safer wording states only what the numbers show, such as that sales and ad spend both rose Oct to Dec, and that last year's sales rose without extra spend, without claiming ads cause the increase"]}