{"id":"mkt2-test-status-table-planned-sample","audience":"product-managers","category":"data","track":"marketing","technique":"verification","level":"intermediate","title":"Mark four A/B tests ready or still running by planned sample","free":false,"task":"Your growth team runs four A/B tests and the weekly meeting needs a status. Write a prompt for a markdown table with columns Test | Reached | Status, where Reached is current visitors per variant as a percent of the planned visitors (rounded to a whole number) and Status is \"Ready to call\" only if Reached is 100% or more, otherwise \"Keep running\". End with a line starting \"Ready:\" like \"Ready: 2 of 4\". Under 80 words.","materialLabel":"Test tracker (attached to your prompt automatically)","material":"A/B TEST TRACKER - Thursday\n\nTest | Planned visitors per variant | Current visitors per variant | Note\nT1 Pricing headline | 10,000 | 10,400 | none\nT2 Signup button color | 8,000 | 5,200 | none\nT3 Onboarding email | 6,000 | 6,000 | none\nT4 Checkout trust badge | 12,000 | 9,000 | paused for 3 days during a site outage\n\nReminder: Tests are only called once they reach their planned sample. Early leads don't count.\nOn T2 the new color is up 12% in clicks; the designer is keen to ship. On T4 the badge variant leads by 0.4 points in conversion.\nNobody has started a fifth test.","rules":[{"type":"plainText"},{"type":"maxWords","max":80}],"mistakes":[{"title":"Calling tests early","why":"The AI lets the 12% lead on T2 drive a Ready call.","fix":"Say: \"Ready to call only if Reached is 100% or more.\"","checks":["r1","r2"]},{"title":"Rounding wrong","why":"The AI shows 104.0% or 65.0% inconsistently.","fix":"Say: \"Whole numbers, with a % sign.\"","checks":["r0"]},{"title":"Forgetting the closing line","why":"The AI ends with commentary.","fix":"Say: \"Last line: Ready: X of 4.\"","checks":["r2"]},{"title":"Letting markdown through","why":"AI adds bold text and headings by default.","fix":"Add: \"Plain text only, no bold, no headings.\"","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and often runs past a word limit.","fix":"Cap each part (for example \"each line under 15 words\") so the total lands well under the limit.","checks":["maxWords"]}],"promptHints":[{"label":"Names the table columns","patterns":["test.*reached.*status"]},{"label":"Defines Reached","patterns":["percent","current .*planned","divided"]},{"label":"Sets the status rule","patterns":["100%","ready to call","keep running"]},{"label":"Gives the final line format","patterns":["ready:","x of 4","2 of 4"]},{"label":"Caps length","patterns":["80 words","plain text","no other text"]}],"hint":"A status table is just rules applied to numbers. Define Reached, the threshold for Ready to call and the final count line.","mission":"Show which of four A/B tests reached their planned sample.","musts":["Ready only at 100%+","Whole-number percent","Ready count line"],"rubricLabels":["The table has the columns Test | Reached | Status and four rows with Reached 104%, 65%, 100% and 75% for T1 to T4","Status is Ready to call for T1 and T3 and Keep running for T2 and T4","The last line is exactly Ready: 2 of 4, and the output does not suggest shipping T2 early"]}