Director, Predictive Modeling

The Coca-Cola Company•Atlanta, GA

1d•Onsite

About The Position

The Predictive Modeling Director is responsible for designing and implementing advanced statistical and predictive models to solve complex analytical problems and guide business decision‑making. In this role, you will be one of the senior data scientists within the Modeling and Measurement team, translating complex business questions into rigorous analytical frameworks and tools that help identify drivers, prioritize investments, and inform strategy. You will focus on developing and refining models that address questions such as optimizing resource allocation, identifying and quantifying drivers of share and growth, understanding customer behavior over time, and forecasting outcomes under different business scenarios. Working closely with cross‑functional partners, you will ensure models are grounded in real business needs and that outputs are clearly communicated and actionable. You will collaborate with a broader team of modelers, analysts, and data engineers, as well as external vendors or agencies as needed, to deliver high‑quality modeling solutions. Ultimately, your work will enable data‑driven planning, smarter prioritization, and improved ROI by providing a predictive lens on key business strategies.

Requirements

Bachelor’s degree in Statistics, Applied Mathematics, Computer Science, Engineering, or another field with a strong quantitative focus.
Demonstrated academic or professional training in advanced statistical modeling and machine learning techniques.
8-10+ years of experience in data science, marketing analytics, or a similar function, with direct experience building predictive models that inform business strategy.
Experience within the CPG or marketing analytics consulting industry is highly valued, especially if you have worked on marketing mix models or consumer predictive analytics.
A track record of translating complex modeling results into actionable business recommendations is required.
Advanced proficiency in statistical analysis and modeling tools.
Strong programming skills in languages like Python or R for data analysis and model development, experience with statistical libraries (pandas, scikit-learn, statsmodels in Python; or equivalent in R).
Deep knowledge of regression analysis, time-series forecasting, machine learning algorithms (regression, classification, clustering, tree-based models, etc.) as applied to marketing problems.
Comfortable working with large datasets; able to write SQL to extract and manipulate data.
Familiarity with data visualization and BI tools (Tableau, Power BI) to present model findings.
Ability to quickly learn and use specialized analytics software & datasets (for example, Nielsen and NielsenIQ or third-party MMM tools, if used) and to evaluate their output critically.
Strong ability to connect analytical work to business outcomes.
Comfortable applying modeling techniques to a range of business problems, including but not limited to marketing, growth strategy, portfolio optimization, forecasting, and resource allocation.
Able to contextualize results within broader strategic and operational considerations.
Excellent problem-solving skills with the ability to frame ambiguous marketing questions into concrete analytical tasks.
Attention to detail to catch data issues or anomalies in model results, combined with a big-picture mindset to focus on insights that matter.
Demonstrated capacity to not just produce data outputs, but to also generate insights – identifying the “so what?” and “now what?” from analysis.
Creativity in analytical approach, finding ways to measure things that aren’t directly measurable by making assumptions or using proxy data, while clearly stating limitations.
Ability to explain complex analytical concepts to non‑technical audiences in a compelling way.
Strong communication skills, both written (presentations, documentation) and verbal, to act as a bridge between the data science world and marketing teams.
Experience presenting findings or recommendations to marketing or business leaders is required; must be able to defend your methodologies while also being receptive to feedback and questions.
Collaborative working style – open to input from others and adept at working in cross-functional teams.
Comfortable managing multiple stakeholders and projects, with strong project management and prioritization abilities to meet deadlines in a fast-paced environment.
Self-motivated and proactive in identifying opportunities where modeling can add value.
Takes ownership of projects and follows through on commitments.
Demonstrated leadership potential, whether through formally managing team members or informally guiding peers.
Skilled at mentoring junior analysts, giving constructive feedback and fostering growth.
A passion for continuous learning in the analytics field, staying up-to-date with new methods or tools (for example, exposure to AI tools for predictive analytics or new data sources for modeling).
High ethical standards regarding data privacy and responsible use of data in modeling.
All persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification form (Form I-9) upon hire.
Must be currently authorized to work in the United States on a full-time basis and must not require The Coca-Cola Company's sponsorship to continue to work legally in the United States.

Nice To Haves

A Master’s degree in Data Science, Analytics, Economics, or MBA with analytics specialization is preferred.
Leadership experience (formal or informal) such as project lead or people manager for analysts/modelers is preferred, indicating ability to manage projects and mentor others.
Experience presenting findings or recommendations to marketing or business leaders is required; must be able to defend your methodologies while also being receptive to feedback and questions.
While this role does not directly manage a team, it plays an important role in raising the overall quality, consistency, and impact of modeling work across the organization.

Responsibilities

Design and develop statistical and machine‑learning models that tackle high‑impact analytical questions across the business, including marketing mix modeling (MMM), driver analysis, demand forecasting, scenario modeling, customer lifetime value estimation, churn and retention analysis, segmentation and clustering, and other econometric or predictive use cases.
Manage the end‑to‑end modeling process: data sourcing, preprocessing, feature engineering, model selection, validation, interpretation, and iteration.
Ensure models are robust, explainable, and decision‑ready, providing not just predictions but insight into underlying drivers and sensitivities.
Develop tools or frameworks that allow users to explore scenarios and translate model results into clear recommendations.
Collaborate with technology, analytics, or platform teams to integrate modeling outputs into dashboards, planning tools, or recurring reporting processes.
Oversee analytical contributions from external vendors or consultants, ensuring their methodologies align with internal quality standards and business objectives.
Engage closely with cross‑functional partners across Marketing, Human Sciences, IMX, and other business teams to understand decision contexts and define the problems modeling can help solve.
Regularly partner with business leaders and working teams to understand what decisions they are trying to inform.
Ensure each modeling effort is tied to a clear use case and that assumptions, limitations, and implications are clearly communicated.
Collaborate with stakeholders to interpret results and identify how insights can be applied in practice.
Maintain high standards of data integrity, statistical rigor, and methodological transparency across all modeling work.
Partner with Data Engineering, Data Science, and IT teams to access, validate, and prepare data from multiple sources.
Apply best practices to avoid bias, overfitting, and misinterpretation, including appropriate validation techniques, sensitivity analysis, and business reasonableness checks.
Document modeling approaches and assumptions thoroughly and contribute to shared model and code repositories for reproducibility.
Monitor model performance over time and refresh or adapt models as new data, tools, or business conditions emerge.
Translate complex analytical results into clear, compelling insights for non‑technical audiences.
Synthesize findings into focused narratives that highlight implications, tradeoffs, and recommended actions.
Create presentations and visualizations that communicate both the “what” and the “so what,” tailoring the level of detail to the audience.
Present results to senior leaders and cross‑functional teams, providing decision support by outlining options informed by modeling output.
Build trust in predictive models by explaining methodologies clearly and engaging openly with questions and feedback.
Operate as a highly collaborative senior individual contributor within the Modeling and Measurement team.
Partner closely with other modelers, analysts, and analytics leaders, sharing expertise, reviewing work, and contributing to team‑wide standards and best practices.
Provide mentorship and thought leadership by supporting peers on complex analytical challenges, contributing to technical discussions, and sharing learnings across the broader analytics community.
Raise the overall quality, consistency, and impact of modeling work across the organization.