Describe a time you had to influence a senior cross-functional leader to change a launch plan based on ambiguous A/B test results. Be specific: the decision stakes, your hypothesis, how you defined success metrics and guardrails, how you handled disagreement (e.g., disagree and commit), the data artifacts you produced (PR/FAQ, doc, dashboard), the trade-offs you highlighted, and the measurable outcome. What would you do differently next time to raise the bar?