Senior Marketing Scientist

Responsibilities

• Architect, validate, and scale Bayesian/frequentist Marketing Mix Models (MMM) to establish true omni-channel media effectiveness and marginal ROAS curves.
• Design, execute, and evaluate incrementality frameworks, including Matched-Market/Geo-Experiments, Synthetic Control Methods, and randomized controlled trials (RCTs).
• Drive methodology for triangulating MMM outputs, attribution models, and experimental results into unified target response curves.
• Build stochastic forecasting algorithms and lifetime value (LTV) models to optimize budget allocation across portfolio channels.
• Develop reproducible, scalable modeling pipelines in Python or R, maintaining high standards for code quality, version control (Git), and data governance.
• Custom-build adstock, saturation, and decay functions tailored to unique client buying cycles and macro-economic signals.
• Collaborate with data engineers to integrate client data warehouses (Snowflake, BigQuery) with Relay, Mile Marker's open-powered marketing stack.
• Bridge the gap between complex quantitative outputs and executive decision-making—translating statistical outputs into clear, actionable budget reallocations.
• Partner with client strategy leads to draft data-driven Learning Agendas that systematically test incrementality hypotheses.
• Serve as the subject matter expert in client meetings and pitch presentations, building credibility through methodological clarity and mathematical rigor.

Requirements

• 5+ years of professional experience in Marketing Science, Quantitative Marketing, Data Science, or Econometrics, ideally within a high-growth agency, adtech, or direct-to-consumer environment.
• Degree in a quantitative discipline (Statistics, Econometrics, Applied Mathematics, Computer Science, Economics, or related quantitative field) preferred.
• Expert-level mastery of Marketing Mix Modeling (MMM), Bayesian statistics, time-series analysis, and causal inference techniques (Matched Markets, Synthetic Controls, Difference-in-Differences).
• Advanced proficiency in Python (Pandas, NumPy, PyMC, Statsmodels) or R (Tidyverse, Prophet, CausalImpact) for data manipulation, statistical modeling, and automation.
• Strong proficiency in SQL for querying complex relational databases and cloud platforms (BigQuery, Snowflake, Redshift).
• Standard workflows using Git/GitHub .
• Ability to build programmatic visualizations and clear analytical dashboards using tools like Streamlit, Dash, Tableau, or Looker.
• Demonstrated ability to explain complex statistical models, confidence intervals, and incrementality tests to non-technical client executives.
• Proven track record of translating model outputs into clear budget optimization strategies that increase client ROI and enterprise value.