Data Scientist - Media Mix Modeling & Experimentation

Overview

Position Summary...At Sam’s Club, we’re member obsessed — and data powered. The Measurement, Insights & Data Strategy (MINDS) team sits at the center of Sam’s Club’s Member Access Platform (MAP), enabling suppliers, merchants, and marketers to make smarter decisions through trusted measurement, experimentation, and data science.

As a Staff Data Scientist on the Media Science & Experimentation team, you will play a pivotal role in shaping how Sponsored Product Ads (SPA) are measured, optimized, and scaled. This is a hands-on, highly technical individual contributor role where you’ll design and productize advanced measurement, machine learning, and causal inference solutions that directly influence how top advertisers plan, test, and grow their media investments.

Your work will power closed-loop, omnichannel measurement at scale — turning Sam’s Club’s unique member journey data into actionable, revenue-driving insights for both suppliers and the business.

• Build and scale advertiser-ready Media Mix Models (MMM) Design, productize, and deploy granular Sponsored Product Ads MMM solutions that advertisers actively use for budget planning, scenario simulation, and optimization. Establish MMM as a core FY27 Joint Business Plan (JBP) measurement capability with strong adoption and credibility.
• Advance incrementality and causal measurement for Sponsored Ads Lead the design and analysis of A/B tests and quasi-experimental frameworks to quantify true incremental lift from media investments. Move the organization beyond pre/post reporting toward trusted, decision-grade measurement that informs product launches and advertiser strategy.
• Develop machine learning models that improve advertiser growth and member experience Build advertiser-level segmentation and retention models that drive supplier growth, along with member-level ad response prediction models that optimize performance while protecting the member experience.
• Own experimentation strategy for new ad products and formats Partner with Product, Engineering, and Ads teams to design rigorous experiments for SPA product launches, interpret results, and translate findings into clear recommendations that influence roadmap and go-to-market decisions.
• Productize data science solutions at scale Deliver production-grade modeling systems with robust pipelines, monitoring, and retraining. Apply AI-driven approaches to automate workflows, accelerate insights, and ensure measurement solutions are scalable, testable, and repeatable.
• Act as a technical thought partner across MAP Lead cross-functional discussions on measurement and modeling tradeoffs, educate stakeholders, and provide clear, actionable guidance that balances analytical rigor with business impact.

• A master’s degree or higher in Computer Science, Machine Learning, Statistics, Mathematics, Operations Research, or a related quantitative field, plus 5+ years of industry experience applying advanced analytics and machine learning to real-world business problems.
• A proven track record of building and scaling production-grade data science systems, including model training pipelines, evaluation, monitoring, and retraining for decision-critical use cases.
• Deep hands-on experience with Media Mix Modeling (MMM) for budget planning and optimization, including saturation curves, diminishing returns, marginal ROAS, and Bayesian or frequentist approaches — with the ability to translate outputs into clear business recommendations.
• Strong expertise in causal inference and experimentation, including A/B testing, power analysis, difference-in-differences, matched or synthetic controls, and Bayesian time series methods, with at least 2+ years of hands-on causal ML or experimentation experience.
• Solid grounding in machine learning and statistical modeling, including advertiser segmentation, ad response prediction, uplift modeling, and bias–variance tradeoffs; exposure to keyword or bidding optimization is a plus.
• High proficiency in Python and SQL, with experience working in large-scale data environments (e.g., Big Query, Dataproc), distributed data processing, and feature pipeline…