GEICO is looking for a Staff Data Scientist, Product Analytics who will provide quantitative rigor, behavioral insight, and a strategic perspective to partners across the organization. As a curious and decision-oriented member of the team, you serve as the analytical thought partner to product, engineering, and design leaders—using data, experimentation, and causal reasoning to help them make better decisions. You will frame the right questions, design the studies that answer them, and translate findings into recommendations that shape what we build. You'll make critical recommendations for Product, Engineering, and senior leadership.

Job Responsibilities:
•     Strategic Partnership: Embed with product, engineering, and design leaders as a decision partner—framing ambiguous business questions, pressure-testing roadmap assumptions, and shaping bets before they are committed.
•     Metric Frameworks: Define goal metrics, guardrails, and supporting metric trees for your product area. Decompose top-line outcomes into measurable inputs that teams can move.
•     Experimentation: Design, power, and analyze A/B and quasi-experiments. Go beyond average treatment effects to understand heterogeneity, long-term impact, novelty effects, and cross-surface interactions.
•     Causal Inference: Apply causal methods (difference-in-differences, synthetic control, instrumental variables, propensity scoring, switchback designs) where randomization is not feasible.
•     Decision Modeling: Build opportunity sizing, forecasting, and ROI models that scale how the organization makes trade-offs—pricing, growth, retention, and long-range investment decisions.
•     Deep Dives: Lead root-cause investigations into user behavior, funnel performance, retention, and engagement. Translate messy signal into clear, defensible recommendations.
•     Collaboration: Serve as trusted analytics partner to PMs, designers, and engineers—translate product questions into research designs, analyses, and deliverables.
•     Communication: Present findings and recommendations to senior leadership with clarity, structure, and its uncertainty.
•     Mentorship & Best Practices: Mentor junior data scientists and analysts. Define internal standards for experiment design, statistical rigor, and reproducible analysis.

Basic Qualifications:
•     Experience: 8+ years in product analytics, decision science, data science, or a closely related quantitative role at a technology company, with a track record of influencing product decisions.
•     Experimentation: Strong background designing well-powered A/B tests, diagnosing bias and variance issues, handling interference and non-stationarity, and interpreting results under real-world conditions.
•     Statistics & Causal Inference: Solid applied statistics foundation with practical experience selecting the right causal method for the question at hand.
•     Tools: Advanced SQL and working proficiency in Python or R for analysis, modeling, and reproducible workflows.
•     Product Sense: Demonstrated ability to define metric frameworks for a product area, not just operate within frameworks built by others.
•     Communication: Clear writing and direct storytelling with data—can produce a one-pager that moves a roadmap and present to executives without losing the nuance.
•     Quality Mindset: Disciplined about quantifying what you know, surfacing what you don't, and resisting false precision.
•     Education: Bachelor's degree or higher in statistics, economics, computer science, mathematics, operations research, or a related quantitative field—or equivalent practical experience.

Preferred Qualifications:
•     Advanced Methods: Bayesian methods, hierarchical models, sequential testing, or uplift modeling experience.
•     Domain Experience: Background in growth, pricing, monetization, marketplace, or recommendation problem spaces.
•     ML Partnership: Experience working alongside ML engineers on production models, including offline evaluation, online metrics, and guardrails.
•     Platform Contributions: Contributions to internal experimentation platforms, metric stores, or measurement tooling.
•     Education: MS or PhD in a quantitative discipline.
•     Industry: Insurance or financial services experience.