{"id":"cx-escalation-matrix-four-cases","audience":"customer-support","category":"support-work","technique":"grounding","track":"support","level":"intermediate","title":"Route four cases using the escalation matrix","free":false,"task":"Route each case using the matrix. Output exactly four lines starting \"Case 1:\" to \"Case 4:\", each formatted like \"Case 1: Role (Name)\". If more than one row matches, use the row listed first. Plain text, nothing else.","materialLabel":"Your material (attached to your prompt automatically)","material":"ESCALATION MATRIX (use the first matching row)\n1. Possible security issue or data exposure -> Security on-call (Sam Ito)\n2. Platform-wide outage -> Incident manager (Ben Ruiz)\n3. Customer mentions a lawyer or legal action -> Legal (Nora Platt). Support does not give legal opinions.\n4. Customer contract worth more than $10,000 per year at risk -> Account director (Priya Nand)\n5. Anything else -> Team lead (Tess Moore)\n\nCASES\nCase 1: A customer reports that reports from another company appear in their dashboard.\nCase 2: A customer with a $45,000 contract says they will cancel unless a billing error is fixed this week.\nCase 3: A customer writes that their lawyer will contact us about a contract dispute.\nCase 4: The login page fails for every user in three countries.\n\nNote: Case 4 customers include several large accounts.","rules":[{"type":"plainText"},{"type":"maxWords","max":50}],"mistakes":[{"title":"Choosing the revenue row for the outage","why":"Several large accounts are mentioned, so the AI picks the account director.","fix":"Say: \"Use the first matching row.\"","checks":["r2"]},{"title":"Giving advice instead of routing","why":"The AI writes what the customer should be told.","fix":"Say: \"Output only Role (Name) per case.\"","checks":["r0"]},{"title":"Wrong names","why":"The AI uses a role without the name or mixes names.","fix":"Say: \"Format Role (Name) exactly as in the matrix.\"","checks":["r1"]},{"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":"Trusting the AI to count words","why":"AI is bad at counting and usually runs long, so a vague \"keep it short\" lands well over the limit.","fix":"Cap each part instead of the total, for example \"each line under 15 words\".","checks":["maxWords"]}],"promptHints":[{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Sets a length limit","patterns":["\\d+\\s*words","word (limit|count)","under \\d+"]},{"label":"Gives the line format","patterns":["case 1:","role \\(name\\)"]},{"label":"States the tie rule","patterns":["first matching","first row","listed first"]}],"hint":"Routing tables only work with a tie rule. Give the output format and say which row wins.","rubricLabels":["The output is exactly four lines starting Case 1: through Case 4:, each formatted Role (Name)","Case 1 is Security on-call (Sam Ito) and Case 2 is Account director (Priya Nand)","Case 3 is Legal (Nora Platt) and Case 4 is Incident manager (Ben Ruiz), not Account director"]}