{"id":"cs-health-score-math-for-vp","audience":"customer-success","category":"support-work","technique":"verification","track":"support","level":"intermediate","title":"Explain an account's health score to your VP with the math shown","free":false,"task":"Your VP of Customer Success, Marta, asked why Copperline Insurance shows as Yellow. Recalculate the score using the formula in the notes and the corrected seat count, then write exactly 4 plain-text lines for her: one per component (Logins, Features, Tickets) showing the math, then a last line starting \"Total:\" with the score and band. Under 90 words.","materialLabel":"Account notes (attached to your prompt automatically)","material":"Health score formula (CS ops wiki):\n- Logins (40%): weekly active users / licensed seats\n- Features (30%): core features in use / 10 core features\n- Tickets (30%): 100% if fewer than 5 open tickets, 50% if 5-10 open, 0% if more than 10\n- Bands: Green 60-100, Yellow 40-59, Red under 40\n\nCopperline Insurance - export from Nov 3:\n- Licensed seats: 200\n- Weekly active users (last 4 wks avg): 120\n- Core features in use: 6 of 10\n- Open tickets: 7 (2 are waiting on customer)\n- Score shown in dashboard: 57 (Yellow)\n- NPS last survey: 31\n\nNote from billing, Nov 4: Copperline's seat count dropped from 200 to 160 at their Oct 1 mid-term change. The dashboard still pulls the old contract number until the sync runs Friday.\n\nCSM note: Champion (Ade Okafor) mentioned they're piloting a competitor tool in claims. Not confirmed.","rules":[{"type":"maxWords","max":90},{"type":"plainText"}],"mistakes":[{"title":"Trusting the dashboard number","why":"The AI repeats the 57 already shown instead of redoing the math with the corrected seats.","fix":"Say: \"Recalculate with 160 seats from the billing note; the dashboard is out of date.\"","checks":["r0","r2"]},{"title":"Hiding the math","why":"The AI gives a verdict without the steps, so Marta can't check it.","fix":"Ask: \"Show each component like: Logins: 120 / 160 = 75% x 0.4 = 30.\"","checks":["r1"]},{"title":"Adding unrelated signals","why":"The AI pulls in NPS and the competitor rumor to sound thorough.","fix":"Add: \"Use only the formula's inputs. Leave out NPS and rumors.\"","checks":["r2"]},{"title":"Trusting the AI to count words","why":"AI explains each step at length and runs long.","fix":"Cap each line: \"each line under 18 words\", well under 90 total.","checks":["maxWords"]},{"title":"Letting markdown in","why":"The AI bolds the total and adds a heading.","fix":"Say: \"Plain text, no bold or headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Uses the corrected seat count","patterns":["160","corrected","billing note","seat"]},{"label":"Asks to show the math","patterns":["show.{0,15}(math|work|calculation)","x 0\\.4","weight"]},{"label":"Sets the line count","patterns":["\\b(4|four)\\b.{0,10}lines?"]},{"label":"Says who reads it","patterns":["marta","vp"]},{"label":"Sets a word limit","patterns":["\\d+\\s*words"]}],"hint":"Marta wants to check the number herself. Make the AI redo the math with the corrected input and show every step.","rubricLabels":["Logins uses the corrected 160 seats: 120 / 160 = 75%, weighted to 30 points (not 200 seats or 24 points)","Features shows 6/10 = 60% for 18 points and Tickets shows 7 open = 50% for 15 points, each with its math on its own line","The last line starts with 'Total:' and gives 63 and Green (not 57 or Yellow), and the reply leaves out the NPS and the competitor rumor"]}