Description
This role will partner closely with external data vendors and media partners, lead geo-experimentation design and measurement, and develop propensity and optimization models that guide media investment decisions. We value continuous learning, rigorous causal thinking, and the ability to translate complex analysis into clear, actionable recommendations for marketing leaders.
Responsibilities
- Marketing Mix Modeling: Serve as the internal analytical counterpart to our Marketing Mix Modeling vendor, reviewing quarterly model refreshes for accuracy, geographic granularity, variable specification, and channel decay assumptions, and reconciling MMM outputs against internal experimental results. Build MMM in-house to measure the return on investment (ROI) across different advertising channels. Translate MMM and incrementality outputs into media/channel budget allocation recommendations, working with internal stakeholders to connect model output to planning cycles
- Experimentation & Measurement: Design and analyze geo-experiments, media heavy-up/heavy-down designs, and matched-market or synthetic control tests across channels, including market selection, statistical power/MDE calculations, and exclusion protocols for confounding risk e.g., seasonal exposure, outlet share, promotions, etc.
- Build Optimization Frameworks: Design randomized holdout structures to measure incremental value of Email/SMS programs, and develop send-time, frequency, and content optimization models in partnership with other marketing entities
- Develop Propensity & Customer Models: Build propensity-to-purchase, churn, and channel-response models using transactional, loyalty, and behavioral data to support targeting, personalization, and customer segmentation
- Analyze Trends & Validate Data: Interpret customer and media performance data for actionable patterns, assess new data sources for improving model(s) accuracy, and maintain rigorous data validation standards
- Collaborate for Implementation: Partner with internal teams in advanced analytics work to operationalize models and ensure measurement infrastructure scales
Qualifications
- Experience: 3–5 years in data science, marketing analytics, or applied statistics, ideally with direct exposure to marketing measurement in a retail, or omnichannel environment
- Technical Skills: Proficiency in Python and/or R and SQL; hands-on experience with causal inference methods, Bayesian modeling, experimental design and power analysis, and marketing mix modeling concepts. Familiarity with Excel, cloud data warehouses and BI/visualization tools (Tableau, Looker, GCP, CDP)
- Dynamic Environment: Comfort working in a research-oriented group with multiple concurrent projects
- Communicate for Impact: Ability to translate statistical and modeling work into clear, decisionready recommendations for non-technical executive stakeholders
- Educational Background: Bachelor's degree in statistics, information systems, data science, applied mathematics, economics, computer science, or related discipline; Master's degree preferred