Scalable bundling via dense product embeddings
This work addresses the practical challenge of scalable bundle design for retailers, offering a data-driven approach that improves upon traditional theoretical assumptions, though it is incremental in applying embeddings to a known problem.
The authors tackled the problem of designing product bundles in retail by developing a machine-learning methodology that uses dense product embeddings derived from purchase and clickstream data, and found that their embedding-based heuristics strongly predict bundle success, with robust performance across categories and generalization to the entire assortment.
Bundling, the practice of jointly selling two or more products at a discount, is a widely used strategy in industry and a well examined concept in academia. Historically, the focus has been on theoretical studies in the context of monopolistic firms and assumed product relationships, e.g., complementarity in usage. We develop a new machine-learning-driven methodology for designing bundles in a large-scale, cross-category retail setting. We leverage historical purchases and consideration sets created from clickstream data to generate dense continuous representations of products called embeddings. We then put minimal structure on these embeddings and develop heuristics for complementarity and substitutability among products. Subsequently, we use the heuristics to create multiple bundles for each product and test their performance using a field experiment with a large retailer. We combine the results from the experiment with product embeddings using a hierarchical model that maps bundle features to their purchase likelihood, as measured by the add-to-cart rate. We find that our embeddings-based heuristics are strong predictors of bundle success, robust across product categories, and generalize well to the retailer's entire assortment.