{"id":"cs-outage-message-three-segments","audience":"managers","category":"support-work","technique":"audience","track":"support","level":"intermediate","title":"Tailor one outage message for three customer segments","free":false,"task":"Write the same outage message for three segments as exactly three lines starting \"Enterprise:\", \"Mid-market:\" and \"Self-serve:\". Each is under 28 words and follows its segment rule. All say the issue started at 10:05 AM and the next update is at 12:00 PM Eastern. Plain text, no emojis.","materialLabel":"Your material (attached to your prompt automatically)","material":"OUTAGE (Nov 6): the reporting feature has been unavailable since 10:05 AM Eastern. Next update at 12:00 PM Eastern. No data is lost.\n\nSEGMENT RULES\n- Enterprise: addressed by the named contact approach: mention that their account manager will call them before noon, and give the phone line 555-0111.\n- Mid-market: email style: mention that they can reply to this email with questions.\n- Self-serve: short, point to the status page at status.example.test; no phone offer, no reply offer.\nAll segments: no credits mentioned; no cause given (still being investigated).\nOld message (retired): 'We are experiencing an issue due to a database fault.' (Cause is unconfirmed.)","rules":[{"type":"plainText"},{"type":"maxWords","max":100}],"mistakes":[{"title":"Giving the cause","why":"The old message names a database fault; the cause is unconfirmed.","fix":"Say: \"No cause, no credits.\"","checks":["r2"]},{"title":"Same message for everyone","why":"The AI copies one text three times.","fix":"Say: \"Follow each segment rule: call, reply, status page.\"","checks":["r1"]},{"title":"Wrong layout","why":"The AI writes a heading and a long text.","fix":"Name the three labels.","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 segment labels","patterns":["enterprise:","mid-market:","self-serve:"]},{"label":"Bans cause and credits","patterns":["no cause","no credits"]}],"hint":"Segmented messages share facts but differ in channel and tone. Name the segments, their rules and what to leave out.","rubricLabels":["The output is exactly three lines starting Enterprise:, Mid-market: and Self-serve:, each saying the issue began at 10:05 AM and the next update is at 12:00 PM Eastern","Enterprise mentions the account manager calling before noon and 555-0111, Mid-market invites a reply to the email, and Self-serve points to status.example.test with no phone or reply offer","No line gives a cause (such as a database fault) or mentions credits"]}