{"id":"cs-nudge-campaign-results-honest-summary","audience":"managers","category":"support-work","technique":"verification","track":"support","level":"intermediate","title":"Summarize a nudge campaign without claiming it caused the lift","free":false,"task":"Write the summary for the VP in exactly 3 bullets starting \"-\": \"Result:\", \"Caution:\" and \"Next step:\". Result shows both rates as counts. Caution says why this does not prove the campaign worked. Do not use the word \"caused\" or claim impact. Plain text, under 70 words.","materialLabel":"Your material (attached to your prompt automatically)","material":"ADOPTION NUDGE CAMPAIGN - September (feature: scheduled reports)\n- 12 accounts received the 4-email nudge series. CSMs chose these accounts because they thought the admins would respond.\n- 12 comparable accounts (similar size and plan) did not receive it.\n- At day 30: 7 of the 12 nudged accounts were using scheduled reports; 5 of the 12 not nudged were.\n- 3 of the nudged accounts also had a training session in the same month.\n- Campaign cost: about 6 hours of CSM time.\n\nVP's question: \"Did the campaign work? Can we roll it out to everyone?\"\nA CSM commented: \"It obviously worked, 58% vs 42%.\"","rules":[{"type":"plainText"},{"type":"maxBullets","max":3},{"type":"maxWords","max":70}],"mistakes":[{"title":"Claiming it worked","why":"The AI agrees with the CSM and says the campaign lifted adoption.","fix":"Say: \"Do not claim impact; explain why this does not prove it.\"","checks":["r2"]},{"title":"Only percentages","why":"The AI writes 58% vs 42% without the small counts.","fix":"Say: \"Show both rates as counts: 7 of 12 and 5 of 12.\"","checks":["r1"]},{"title":"No next step","why":"The AI ends with a rollout recommendation.","fix":"Say: \"Next step: test on randomly chosen accounts.\"","checks":["r2"]},{"title":"Getting markdown back","why":"AI adds bold and headings by default, which show up as stray symbols where this is pasted.","fix":"Say \"plain text, no bold, no headings, no asterisks\".","checks":["plainText"]},{"title":"Leaving the bullet count open","why":"The AI keeps adding bullets until every detail is covered.","fix":"Name the cap: \"at most N bullets, nothing else\".","checks":["maxBullets"]},{"title":"Trusting the AI to count words","why":"The AI runs long and cannot count its own words.","fix":"Cap each part (\"each line under 12 words\") and add \"under N words in total\".","checks":["maxWords"]},{"title":"Leaving a requirement unstated","why":"The AI follows its own defaults unless the prompt spells this out.","fix":"State it plainly in your prompt: the output has exactly three bullets starting '- Result:', '- Caution:' and '- Next step:'.","checks":["r0"]}],"promptHints":[{"label":"Names the three bullets","patterns":["result","caution","next step"]},{"label":"Uses counts","patterns":["7 of 12","counts","of 12"]},{"label":"Avoids causal claims","patterns":["caused","claim","prove","impact"]},{"label":"Asks for a random test","patterns":["random","test"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Caps the bullets","patterns":["bullet"]},{"label":"Sets a word cap","patterns":["\\d+\\s*words","under \\d+","word"]}],"hint":"Small, hand-picked samples can mislead. Tell the AI to show counts, say why it proves nothing and propose a fair test.","mission":"Summarize the nudge campaign for the VP without overclaiming.","musts":["Show rates as counts","Explain why no proof","Never say caused"],"rubricLabels":["The output has exactly three bullets starting \"- Result:\", \"- Caution:\" and \"- Next step:\"","Result gives 7 of 12 nudged versus 5 of 12 not nudged as counts","Caution says the accounts were hand-picked by CSMs (and/or some also had training) so the campaign cannot be shown to have caused the gap, and Next step proposes a test on randomly chosen accounts; the word \"caused\" is not used as a claim"]}