Diagnose 10–11% usage drop across geos

Usage in the US has fallen by 10%, and Mexico by 11%. Enumerate plausible confounders (seasonality, pricing, outages, marketing mix, competitor moves, feature rollouts, macro). Design an experiment to isolate causality: specify the randomization unit (user or geo), exposure and segmentation, primary metric and guardrails, duration and power, and pre-registration. If an experiment is not possible, outline a difference-in-differences or synthetic control approach with explicit identification assumptions and diagnostics. State which covariates to stratify or control for, how you would cut the results, and how you would communicate findings and confidence intervals to stakeholders together with remediation options.

Overview: This question tests causal inference and experimentation skills, including metric diagnosis, confounder identification, experimental design, quasi-experimental reasoning, statistical power considerations, and communicating results in the Analytics & Experimentation area for a Data Scientist position.