{"id":"mgr-x-hiring-scorecard-evidence","audience":"managers","category":"decisions","track":"data","technique":"grounding","title":"Fill in a hiring scorecard with evidence, not vibes","free":false,"task":"You're writing up your interview scorecard for a customer success candidate. You need exactly 4 numbered lines, one per criterion, each with a score from 1 to 4 and one piece of evidence from your notes, then a final 'Recommendation:' line, plain text, under 110 words. Base scores only on job-related evidence; leave out anything personal or about 'culture fit', and follow the hiring rule for the recommendation.","materialLabel":"Your interview notes (attached to your prompt automatically)","material":"Interview notes - Maya Chen, Customer Success Manager, Tue Nov 4\nScale: 1 = no evidence, 2 = some, 3 = solid, 4 = exceptional.\nHiring rule: recommend Hire only if every criterion scores 3 or higher.\n\nCriterion 1 - Handling an upset customer:\nWalked through a churn-risk account at her last job; set up weekly calls, got them to renew at the same price. Clear, specific. Strong.\n\nCriterion 2 - Product learning speed:\nIn the case exercise she learned our reporting feature from the help doc in 10 minutes and explained it back correctly. Solid.\n\nCriterion 3 - Data use:\nSaid she \"looks at health scores\" but couldn't give an example of a decision she made from data. Vague even after a follow-up.\n\nCriterion 4 - Written communication:\nFollow-up email after the case was clear, short, had next steps. Solid.\n\nOther impressions (not criteria):\n- Mentioned she has two young kids and needs to leave at 5 some days.\n- Has an accent, \"might be hard on calls\"? (a panelist said this)\n- Great culture fit, we like the same hiking trails!","rules":[{"type":"numberedItems","count":4},{"type":"maxWords","max":110},{"type":"plainText"}],"mistakes":[{"title":"Asking for 'a write-up'","why":"The AI writes a narrative assessment with headings and a summary of strengths and concerns.","fix":"Be exact: \"Exactly 4 numbered lines, one per criterion, then a Recommendation: line. Plain text.\"","checks":["numberedItems","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 line instead: \"each line under 20 words.\"","checks":["maxWords"]},{"title":"Ignoring the hiring rule","why":"Three strong scores make the AI lean toward Hire.","fix":"Add: \"Apply the rule: Hire only if every criterion is 3 or higher.\"","checks":["r1"]},{"title":"Scoring without evidence","why":"The AI rates on overall impression, and the weak area gets rounded up.","fix":"Add: \"Each line: score, then one piece of evidence from my notes.\"","checks":["r0"]},{"title":"Including personal details","why":"The 'other impressions' are in your notes, and the AI may treat them as signals.","fix":"Add: \"Job-related evidence only. Nothing about family, accent or culture fit.\"","checks":["r2"]}],"promptHints":[{"label":"Asks for evidence","patterns":["evidence","example","from my notes"]},{"label":"Applies the hiring rule","patterns":["rule","3 or higher","every criterion"]},{"label":"Keeps personal details out","patterns":["personal","job.related","culture fit","kids","accent"]},{"label":"Sets the number of lines","patterns":["\\b(4|four)\\b.{0,15}(line|criteri)"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"A fair scorecard is evidence plus a rule. Tell the AI one line per criterion with evidence, how to apply the hiring rule, and what must never count.","rubricLabels":["Scores match the evidence: upset customer 4 (or 3), product learning 3, data use 1 or 2, written communication 3, each with one piece of evidence from the notes","The final line is 'Recommendation:' with No hire (or Do not hire), because data use is below 3 under the hiring rule","Mentions nothing personal or non-job-related (no kids or leaving at 5, no accent, no hiking or culture fit), and has no intro or extra commentary"]}