{"id":"mkt2-sponsorship-decision-using-actuals","audience":"executives","category":"evaluation","track":"marketing","technique":"verification","level":"advanced","title":"Recommend on an event sponsorship using last year's actual leads","free":false,"task":"Your CMO, Joan, must decide on a $15,000 conference sponsorship. Write a prompt for exactly 3 plain-text lines: \"Math:\" with the full cost divided by expected leads based on last year's actual leads from the same event, \"Target:\" with the target CPL and whether the result is above or below it, and \"Recommendation:\" Sponsor or Decline with a reason under 12 words. Under 70 words.","materialLabel":"Sponsorship offer and history (attached to your prompt automatically)","material":"SPONSORSHIP DECISION - DataSummit 2027\n\nSponsorship fee: $15,000 (Gold booth).\nAdditional costs not in the fee: booth staff travel and hotel, about $2,000.\nVendor (the organizer) says: \"Gold sponsors typically collect around 100 leads.\"\nOur actual result at DataSummit 2026: $12,000 total cost, 60 leads collected.\nTarget cost per lead (CPL): $200.\n\nOther benefits mentioned: a logo on the website, a speaking slot (we would have to apply; not guaranteed).\nContext: the CEO likes the event; last year's meetings booked from the 60 leads: 9 (not part of the target).\nDeadline to commit: Nov 20.","rules":[{"type":"plainText"},{"type":"maxWords","max":70}],"mistakes":[{"title":"Using the vendor's 100 leads","why":"The AI uses the organizer's number, giving $170.","fix":"Say: \"Use last year's actual leads (60), not the vendor's estimate.\"","checks":["r0","r2"]},{"title":"Forgetting travel costs","why":"The AI divides $15,000 only.","fix":"Say: \"Full cost includes the $2,000 travel.\"","checks":["r0"]},{"title":"Liking the event","why":"The AI weighs the CEO's enthusiasm and the speaking slot.","fix":"Say: \"Decide only on CPL against the target.\"","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","r1"]}],"promptHints":[{"label":"Names the three labels","patterns":["math:.*target:.*recommendation:"]},{"label":"Uses actual leads","patterns":["actual","60","last year"]},{"label":"Includes all costs","patterns":["full cost","\\$2,000","travel","plus"]},{"label":"Compares to target","patterns":["\\$200","above or below","target"]},{"label":"Caps reason and length","patterns":["under 12 words","70 words"]}],"hint":"Estimates from the seller are not evidence. Tell the AI to use last year's actual leads, include every cost and decide against the target.","mission":"Tell CMO Joan whether to buy the $15,000 sponsorship.","musts":["Use last year's actual leads","Full cost divided by leads","Compare to target CPL"],"rubricLabels":["The Math line divides the full cost of $17,000 ($15,000 plus $2,000) by 60 leads (last year's actual), giving about $283","The Target line gives $200 and says the result is above the target","The Recommendation line says Decline with a reason under 12 words, and the output does not use the organizer's 100-lead estimate"]}