{"id":"data2-anomaly-note-signup-spike-vp","audience":"operations","category":"data","track":"data","technique":"audience","level":"intermediate","title":"Write an anomaly note on a signup spike for a VP","free":false,"task":"Yesterday's signups spiked and the VP of Marketing, Dana Whitfield, wants to know if it is real. Write a short plain-text message to her under 100 words. First sentence: how many signups versus a normal day. Then show the math splitting real-looking signups from bot-pattern ones and what the campaign explains. End with one line starting 'Next:' saying you will re-run the count Wednesday morning. Don't repeat unconfirmed rumors.","materialLabel":"Notes from the data channel (attached to your prompt automatically)","material":"Signups - daily report, Tuesday\n\nNormal weekday: about 300 signups.\nTuesday: 1,150 signups.\n\nFindings so far:\n- The bot filter was switched off for 6 hours on Tuesday (ticket OPS-77), restored at 8 pm.\n- 640 accounts created in that window match the bot pattern (same email domain, created seconds apart).\n- A paid social campaign went live Tuesday morning. The plan expected about 150 extra signups from it.\n- Someone in the channel said a competitor might be scraping our signup form. Nobody has checked this. Treat as unverified.\n- The 640 bot-pattern accounts are not counted as customers in the CRM.\n\nOpen questions:\n- Whether the remaining signups after bots and the campaign are fully explained (not yet known).\n- Whether the filter outage also let through bots that don't match the pattern.\n\nDana's preference: short, numbers first, no jargon.","rules":[{"type":"plainText"},{"type":"maxWords","max":100}],"mistakes":[{"title":"Burying the headline","why":"The AI opens with a greeting and context before the 1,150 vs 300.","fix":"Add: \"First sentence: 1,150 signups vs about 300 normal.\"","checks":["r0"]},{"title":"Skipping the split math","why":"The AI says 'mostly bots' without showing 1,150 - 640 = 510 and the campaign share.","fix":"Add: \"Show 1,150 - 640 = 510, then 510 - 300 - 150 = 60 unexplained.\"","checks":["r1"]},{"title":"Repeating the rumor","why":"The AI includes the scraping idea as a possible cause and omits the next step.","fix":"Add: \"Leave out the competitor rumor. End with Next: re-run Wednesday morning.\"","checks":["r2"]},{"title":"Leaving the format to chance","why":"The AI picks its own length and layout unless told the exact shape.","fix":"Add: \"Plain text, no bold or headings.\"","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting. Ask for a total word limit and it often lands over it.","fix":"Cap each part instead, for example \"each line under 15 words.\"","checks":["maxWords"]}],"promptHints":[{"label":"Names the reader","patterns":["dana","vp"]},{"label":"Asks for the headline first","patterns":["first sentence","1,?150"]},{"label":"Asks to show the math","patterns":["1,?150 ?- ?640","510","show the math"]},{"label":"Rules out the rumor","patterns":["rumor","unverified","unconfirmed","leave out"]},{"label":"Fixes the last line","patterns":["next:","wednesday"]}],"hint":"A good anomaly note shows the arithmetic, separates what's known from what isn't, and leaves rumors out.","mission":"Write a note about the signup spike for the VP.","musts":["Numbers first","Show the split math","End with Next: line"],"rubricLabels":["The first sentence states 1,150 signups versus about 300 on a normal day","The math shows 1,150 minus 640 bot-pattern accounts is 510, and that after about 300 normal and about 150 from the campaign, about 60 signups remain unexplained (510 - 300 - 150 = 60)","The message does not state the competitor-scraping idea as fact or as a finding, and it ends with a line starting 'Next:' about re-running the count Wednesday morning"]}