{"id":"cx-weekly-ticket-trend-readout","audience":"customer-support","category":"support-work","technique":"verification","track":"support","level":"intermediate","title":"Write a weekly ticket-trend readout that sets the outage aside","free":false,"task":"Your VP, Sam, wants this week's ticket trends in exactly four labeled lines: Total:, Biggest rise:, Biggest drop:, Outage:. Show the counts as before and after with the change, like \"60 to 75 (+15)\". Leave the 48 outage-tagged tickets out of every comparison, and report them only on the Outage line. Plain text, under 80 words.","materialLabel":"Your material (attached to your prompt automatically)","material":"TICKET VOLUME BY CATEGORY (helpdesk export)\n\nCategory     Week 41   Week 42\nBilling        120       132\nLogin           85       140\nHow-to         210       180\nBugs            60        75\nShipping        45        46\nTOTAL          520       573\n\nNote: Week 42 Login includes 48 tickets tagged OUTAGE from the Oct 15 login outage (resolved same day, 3 hours).\nWeek 41 had no outage tickets.\n\nAnalyst comment (draft): \"Login tickets jumped 65%, our biggest problem area this week.\"\n\nOther: CSAT 4.3 vs 4.4. Average first reply 3h 10m vs 2h 50m.","rules":[{"type":"plainText"},{"type":"maxWords","max":80}],"mistakes":[{"title":"Counting the outage as a trend","why":"The AI compares raw totals, so a one-day outage looks like a growing problem.","fix":"Say: \"Remove the 48 outage tickets from every comparison.\"","checks":["r0","r1"]},{"title":"Repeating the analyst draft","why":"The AI copies \"Login jumped 65%\" from the comment in the export.","fix":"Add: \"Do your own math. Ignore the analyst comment.\"","checks":["r1"]},{"title":"Leaving the shape open","why":"The AI writes a mini-report with CSAT and reply times Sam didn't ask for.","fix":"Give the 4 labels: \"Total:, Biggest rise:, Biggest drop:, Outage: only.\"","checks":["r2"]},{"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 80 words.","fix":"Cap each part instead of the total: \"each line under 15 words\".","checks":["maxWords"]}],"promptHints":[{"label":"Removes the outage","patterns":["outage","48"]},{"label":"Gives the labels","patterns":["biggest rise","biggest drop","total:"]},{"label":"Shows before and after","patterns":["before","to \\d+","\\(\\+","change"]},{"label":"Says who it's for","patterns":["sam","vp"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"A one-day outage can make a week look worse than it was. Tell the AI to set it aside, show before-and-after counts, and use your four labels.","rubricLabels":["The Total line compares 520 to 525 (+5) once the 48 outage tickets are removed, not 520 to 573","Biggest rise names Bugs, 60 to 75 (+15), and does not name Login","The output is exactly the four lines Total:, Biggest rise:, Biggest drop:, Outage: in that order, with Biggest drop naming How-to, 210 to 180 (-30), and Outage giving the 48 tickets from Oct 15"]}