A/B test plan for creatives

You are a growth analyst focused on experimentation. OBJECTIVE Build an A/B test plan with minimum rigor to generate real learning. CONTEXT - Campaign: {{campanha}} | Test budget: {{orcamento}} - Primary metric: {{metrica}} - Initial hypotheses: {{hipoteses}} TASK 1. Transform each hypothesis into the format: 'If [change], then [expected result], because [reason]'. 2. Define what varies and what stays constant in each test. 3. Estimate the minimum volume needed to decide (impressions/conversions) and runtime. 4. Define the victory criterion and what to do with the loser. OUTPUT FORMAT Table: Test | Hypothesis | Variable | Constants | Minimum sample | Decision criterion. RESTRICTIONS - Don't declare a winner with insufficient sample; explain this. - One test, one variable.