Belk is a retail company seeking an Associate Decision Scientist to support the maintenance and improvement of propensity models that drive customer engagement. The role involves collaborating with analytics and marketing teams to ensure model accuracy and translate data into actionable insights.
Maintain and optimize existing propensity and response models (purchase, churn, reactivation, category affinity, channel, promotion, and creative recommendations) to ensure ongoing accuracy and effectiveness at the individual customer level
Monitor model performance and health, including drift detection, retraining cycles, validation, and continuous improvement of marketing decisioning models
Support model deployment and operations within real-time and near-real-time marketing environments, partnering with Marketing Technology teams to maintain scoring and decisioning capabilities
Execute and analyze A/B, multivariate, holdout, and incrementality tests to evaluate model performance and measure marketing-driven lift
Translate model outputs into actionable marketing strategies by partnering with campaign and CRM teams to develop audience segments, suppression lists, and treatment assignments
Apply and support marketing optimization frameworks that balance short-term revenue objectives with long-term customer lifetime value (CLV) growth
Develop, maintain, and enhance customer-level predictive features and feature stores using transactional, behavioral, engagement, loyalty, and third-party data sources
Analyze customer behavior and segmentation data to deepen understanding of loyalty tiers, shopping occasions, affinities, channel responsiveness, and promotional sensitivity
Develop, maintain, and troubleshoot production-quality analytics solutions, including Python/R model code, feature engineering workflows, and SQL-based data pipelines
Prepare, validate, and manage modeling datasets and scoring processes to support reliable model execution and reproducibility
Document model methodologies, assumptions, performance results, and governance requirements to ensure transparency, compliance, and operational continuity
Communicate model performance and analytical insights to business stakeholders, translating technical findings into clear recommendations while supporting cross-functional collaboration and ongoing professional development
Qualification
Required
Bachelor's Degree in Statistics, Mathematics, Computer Science, Data Science, Economics, or related quantitative field
1-2 years Applied data science or quantitative analytics, with exposure to predictive modeling and machine learning in a business context
Solid proficiency in Python and/or R for statistical modeling, machine learning, and data manipulation
Working knowledge of supervised and unsupervised ML algorithms: gradient boosting (XGBoost, LightGBM), neural networks, clustering, and survival models
SQL skills for complex data extraction and feature engineering from large enterprise datasets
Developing ability to frame business problems into structured analytical approaches, with growing comfort working within existing model designs
Foundational understanding of customer lifecycle economics, CLV modeling, and the mechanics of CRM and loyalty marketing
Basic familiarity with incrementality, experimental design, and the distinction between correlation and causal lift
Ability to communicate quantitative concepts and model results clearly to non-technical stakeholders
Willingness to collaborate cross-functionally and communicate analytical findings clearly to marketing and business partners
Eagerness to learn and grow within a collaborative data science team, with a strong sense of ownership and attention to detail
Ability to manage time and workload effectively with flexibility to shift priorities based on business need
Must be authorized to work in the U.S. without the need for employment-based visa sponsorship now or in the future, this includes OPT
Preferred
Master's Degree in Statistics, Data Science, Operations Research, Machine Learning, or related field
1+ years Working with or supporting customer-level models in a retail, e-commerce, or CRM/loyalty marketing context
Working with models in production environments; familiarity with CDP platforms (e.g., Salesforce Marketing Cloud, Adobe, Braze), a plus
Exposure to or coursework in reinforcement learning, multi-armed bandit, or contextual bandit approaches for real-time decisioning is a plus
Familiarity with cloud-based data environments (Snowflake, Databricks, AWS, GCP); exposure to MLOps or model deployment pipelines is a plus
Benefit
Belk, Inc. is the mainline department store company with 305 stores located in 16 Southern states.
