Whenever you're ready, walk me through how you'd approach this.
How do you determine the required sample size for an experiment before it launches?
Explain how you would translate a business-relevant minimum detectable effect into sample size and test duration. Your response should address the primary metric, guardrails, statistical power, allocation, randomization unit, and how you would prevent peeking or invalid conclusions. Use explicit assumptions and show the calculation for a binary conversion metric.