Product Strategy and High-Context Insight
Partner with Product Managers to influence roadmap priorities through structured problem framing and opportunity sizing.
Drive end-to-end analytical discovery: define questions, shape hypotheses, perform root-cause analysis, and identify second-order effects.
Produce high-context, narrative-driven analysis that shapes product decisions—not just dashboards or metrics reporting.
Identify long-term analytical needs and proactively set the direction for how the product should be measured.
Metrics Architecture and Observability
Own the metric hierarchy for the product domain (north-star, input metrics, counter-metrics, diagnostic metrics).
Design tracking plans and instrumentation schemas; work with engineering to ensure correct implementation.
Implement automated alerting and monitoring to surface health issues, funnel breaks, outliers, and behavior shifts in real time.
Experimentation and Causal Inference
Lead the design, execution, and interpretation of experiments (A/B tests, multivariate tests, staggered rollouts).
Select and apply appropriate statistical techniques (frequentist/Bayesian, CUPED, time-series methods).
Evaluate experiment readiness, calculate required sample sizes, and estimate expected effects.
Translate experiment outcomes into clear recommendations for product and business stakeholders.
Analytical Leadership and Collaboration
Mentor and support junior analysts; set standards for analytical depth, clarity, and rigor.
Act as the analytical owner in cross-functional initiatives, ensuring alignment on metrics and success criteria.
Influence how PMs and engineers adopt analytical best practices, instrumentation, and metric stewardship.
Communicate trade-offs, risks, and insights to both technical and non-technical audiences with precision.
Ownership of Analytical Assets
Partner with Analytics Engineering to ensure models, marts, and pipelines meet quality standards and SLAs.
Ensure the product area has a batteries-included analytical environment—well-documented, reliable, and easy to use.
Maintain data lineage, definitions, and documentation to minimize ambiguity and reduce analytical debt.