{"id":"revops-lead-scoring-rubric-json","audience":"revops","category":"sales-work","technique":"constraints","track":"sales","level":"advanced","title":"Draft a lead-scoring rubric as a JSON config","free":false,"task":"Marketing Ops will paste your lead-scoring draft into their tool, so you need it as one JSON code block: an array of objects with signal, points and reason fields. Use only the signals your analysis found, points must add up to exactly 100, and leave out company headcount because the purchased list it came from is unreliable. After the code block, write one sentence under 25 words asking Sales leadership to review the weights. At most 12 lines of code.","materialLabel":"Lead-to-win analysis (attached to your prompt automatically)","material":"LEAD-TO-WIN ANALYSIS - FY26 inbound leads (n = 4,200, 310 became customers)\nPrepared by RevOps, Oct 2026\n\nSignals that predict a win (win rate when present vs 7% baseline):\n- Requested a demo (not just a whitepaper): 22%\n- Job title is Director or above: 15%\n- Visited pricing page 2+ times in 14 days: 18%\n- Company headcount 200-2,000 (from purchased ZoomData list): 14%\n- Uses a CRM we integrate with: 12%\n\nSignals that did NOT predict a win:\n- Attended a webinar: 7%\n- Opened 3+ marketing emails: 6%\n- Gmail/Yahoo address: 3% (negative signal)\n\nTeam notes:\n- Marketing wants webinar attendance worth 20 points \"because we spend a lot on webinars.\"\n- Current model in the tool gives 10 points per email open.","rules":[{"type":"codeBlocks","count":1},{"type":"maxCodeLines","max":12},{"type":"maxProseWords","max":40}],"mistakes":[{"title":"Not naming the source signals","why":"The AI adds popular scoring signals like webinars and email opens because most models use them.","fix":"Add: \"Use only the signals the analysis found predict a win.\"","checks":["r0"]},{"title":"Missing the headcount problem","why":"The AI scores headcount because it predicts wins in the data. It can't know the list is unreliable.","fix":"Add: \"Leave out company headcount; the purchased list is unreliable.\"","checks":["r0"]},{"title":"No points total","why":"The AI picks round numbers that add up to 85 or 120.","fix":"Add: \"Points must add up to exactly 100. Check the sum.\"","checks":["r1"]},{"title":"Explaining around the code","why":"The AI writes paragraphs before and after the JSON.","fix":"Add: \"After the code block, one sentence under 25 words. Nothing before it.\"","checks":["r2","maxProseWords","codeBlocks"]},{"title":"Long JSON","why":"The AI pretty-prints with every brace on its own line, so the block runs long.","fix":"Add: \"One object per line, at most 12 lines.\"","checks":["maxCodeLines"]}],"promptHints":[{"label":"Asks for JSON","patterns":["json","code block"]},{"label":"Sets the fields","patterns":["signal","points","reason"]},{"label":"Sets the total","patterns":["100","add up","sum"]},{"label":"Drops headcount","patterns":["headcount","purchased list","unreliable"]},{"label":"Limits the prose","patterns":["one sentence","nothing else","review the weights"]}],"hint":"Tell the AI which signals to use, what to leave out and why, that the points total 100, and how much text goes around the code.","rubricLabels":["The JSON includes demo request, Director-or-above title, pricing page visits and CRM integration, and gives no points to company headcount, webinar attendance or email opens","The points in the JSON add up to exactly 100","After the code block there is one sentence asking Sales leadership to review the weights"]}