Mailchimp A/B Test Design and Results Reading Guide
Design a fair test and read results carefully.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it.
Design an A/B test in Mailchimp. Campaign and audience size: [CAMPAIGN_AND_AUDIENCE]. What I want to learn (subject line, send time, call to action): [TEST_GOAL]. Past rates: [PAST_RATES]. Define the hypothesis, one variable to change, the two versions, the split, the success measure chosen in advance (click rate is usually better than open rate; open rates may be unreliable [VERIFY]), and the duration. Explain sample size in plain terms: small differences need large audiences. Example: with 2,000 recipients per version, 40 clicks against 52 clicks is 2.0 percent against 2.6 percent, and the random variation at that size can be large relative to a 0.6 point gap, so it may be noise. Suggest repeating the test or increasing the sample before changing practice. Write a decision rule, avoid peeking early, and do not test many things at once. Provide a log table of tests and results, and record learning even when no winner appears.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.
See an example result (fictional data)
Invented test. Hypothesis: a subject line with a question gets more clicks. Version A 'Your June update' to 2,000 recipients: 40 clicks (2.0 percent). Version B 'Did you see June's offers?' to 2,000: 52 clicks (2.6 percent). Difference 12 clicks, 0.6 points. With these numbers the gap is within normal random variation, so do not declare a winner. Decision rule written in advance: change practice only if a repeat test shows a gap of at least 0.5 points again. Log entry: result inconclusive, retest in July.
Illustrative only, made-up data. Your results will differ and must be reviewed.