Algorithmic Retail Execution. Simple. Scaled.
The challenge
Every day retailers around the world make millions of micro-decisions, and lose profitable revenue because they do not properly leverage AI-based data science for optimal decisions at scale.
Dynamic Choice’s revolutionary algorithms make millions of optimal micro-decisions instantly. Our breakthrough AI-technology is based on more than twenty years of peer reviewed top academic research. Our solutions are designed for large-scale fast-scalable retail applications.
The evidence
Shelf Confidence
The results of our research show:
Availability
80% Online Availability
Root causes
Consumer responses to Stock-Outs
54% switch brand, variant, or cancel
Lost sales
Associate cost
Customer Time Lost
Inventory Record Inaccuracy
Our capabilities
The recent revolution in large language models has shown that one of the biggest breakthroughs in generative AI is the ability to represent noisy and unstructured data, such as text and images. At the core, large scale neural networks can create increasingly complex representations of these data to perform accurate inferences.
Dynamic Choice’s retail execution AI-based technology is a similar breakthrough for representing customer preferences. It builds on more than a decade of pioneering peer-reviewed academic research in top academic journals.
Customers often have complex preferences over products and these preferences are affected by hundreds of factors that are difficult to elicit or understand. Retailers and brands need a compact way to represent these preferences to learn from them.
Dynamic Choice’s customer preference graphs offer this leap forward. It is an interpretable, yet flexible way to represent preferences. Our algorithms jointly analyze customer transaction data with product availability and other signals and compute hundreds of thousands hyper-personalized customer preference graphs.
Dynamic Choice’s technology only needs a few sales transactions and availability records to build a customer preference graph. This is a game-changer because unlike text and images – of which we have an internet’s worth! – retailers often have only very few purchase observations per customer.
Once we have built a database of customer preference graphs, a retailer can use them as a platform for a suite of hyper-personalized applications along the customer journey, such as personalized recommendations, promotions, lists, or advertising. To complete the offering, the team at Dynamic Choice has created proprietary choice models and efficient algorithms that enable sparse implementation of the underlying statistical choice models. Therefore, our technology scales rapidly across thousands of products and customers.
The team
Dynamic Choice is a vibrant New York City start-up. Our co-founders are:
Daniel Corsten
Co-Founder
Daniel is an expert on global retailing and Professor of Operations and Technology at IE University. He was also Visiting Professor at INSEAD, the Wharton School, the London Business School, and the Harvard Business School. Daniel has more than 25 years of experience in retail and consumer goods, as a serial entrepreneur and investor in retail AI startups, and as a Director of the Fulcrum Technology Group. Daniel is a prolific author, most recently of the book ‘Shelf-Confidence’ (with Tom Gruen), the authoritative workbook on retail execution.
Srikanth Jagabathula
Co-Founder
Srikanth is an expert of algorithmic retail execution and Professor of Operations, Technology, and Statistics at New York University’s Stern School of Business. He obtained his PhD from Massachusetts Institute of Technology and was Visiting Professor at Harvard Business School. Srikanth is a serial entrepreneur of retail AI startups, including cofounder of Celect Inc. (acquired by Nike Inc.) and recently worked as Chief Scientific Officer for Arena Ai. Srikanth as more than 15 years of experience in developing AI/ML based algorithms for retail decision making and has won numerous awards for his applied work.
Contact
