{"id":"cx-escalation-note-billing-eng","audience":"customer-support","category":"support-work","technique":"audience","track":"support","level":"intermediate","title":"Write an escalation note a billing engineer can act on","free":false,"task":"Write an escalation note to Jun, a billing engineer who has not seen the ticket. Use exactly four labeled lines: Impact:, Evidence:, Tried:, Ask:. Plain text, under 90 words. Do not include the customer's email, phone or card digits. The Ask line must include a deadline.","materialLabel":"Your material (attached to your prompt automatically)","material":"TICKET 3302 | Priority: High | Account: Harbor & Pine Dental (acct 88-1204)\nContact: Lila Ortiz, lila@harborpine.example, 555-0142\n\nLila (Nov 4): \"You charged me $490 twice for November. Card ending 7710. This is the THIRD time this year and I'm done.\"\n\nAgent 1 (Nov 4, 9:12): Confirmed two charges, ch_7a1 and ch_7a2, both $490.00 on Nov 1, 4 seconds apart. Refunded ch_7a2. Refund shows 'pending'.\nAgent 2 (Nov 5, 8:40): Customer says refund hasn't arrived. Normal bank time is 5-10 business days, told her so. She is furious: first two double charges (Mar, Jul) were also 'fixed' but nobody found the cause.\nAgent 3 (Nov 5, 10:05): Checked her account: she has an auto-retry flag from a failed payment in March that was never cleared. Could be the cause. Cannot clear flags, engineering only.\n\nRenewal for this account is Nov 15.","rules":[{"type":"plainText"},{"type":"maxWords","max":90}],"mistakes":[{"title":"Including private details","why":"The AI copies the contact line, so the note carries the email, phone and card digits.","fix":"Add: \"No email, phone or card digits.\"","checks":["r2"]},{"title":"Missing the likely cause","why":"Agent 3's flag finding sits at the end, so the AI summarizes the anger and drops the lead.","fix":"Say: \"Evidence must include the old auto-retry flag.\"","checks":["r1"]},{"title":"Writing a story","why":"The AI retells the thread instead of using the four labels Jun needs.","fix":"Name the labels and the order: \"Impact:, Evidence:, Tried:, Ask:.\"","checks":["r0"]},{"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":"Names the reader and labels","patterns":["jun","engineer","impact"]},{"label":"Asks for a deadline in the Ask line","patterns":["deadline","by nov","before"]}],"hint":"Jun hasn't seen the ticket. Name the labels, tell the AI which facts must appear, and say what to keep out of the note.","rubricLabels":["Output is exactly four lines labeled Impact:, Evidence:, Tried:, Ask: in that order","Evidence mentions the two $490 charges four seconds apart on Nov 1 and the uncleared auto-retry flag from March, and Tried says one charge was refunded and is pending","Ask line requests that engineering clear the flag and find the root cause, includes a deadline before the Nov 15 renewal, and nothing in the note includes the email, phone number or card digits"]}