How would you A/B test first trade rate?

Coinbase wants to increase **new user first trade rate**. Design an experiment (A/B test) to evaluate a product change intended to increase the probability that a newly signed-up user places their **first trade**. In your answer, cover: 1. **Hypothesis** and what user behavior you are trying to change. 2. **Primary metric** (pick a clear definition, e.g., “first trade within 7 days of signup”) and at least 2 **secondary/guardrail metrics** (risk, compliance, user harm, revenue quality). 3. **Experiment unit & randomization** (user/account/device), eligibility rules, and when a user is considered “exposed”. 4. Key threats to validity (selection bias, learning/novelty effects, interference, instrumentation issues, sample ratio mismatch). 5. **Power/MDE** approach (how you would size the test, what baseline inputs you need). 6. What you would do if the result is **not statistically significant** but directional, or if key guardrails regress.