First, IRI deployed a three-step process to impute a full market view of competitor and client sales, as well as purchase drivers at the store level:
- Data Assembly:
IRI compiled over 100 internal and partner-provided data sets, spanning from point-of-sale and e-commerce information to digital measurement and fuel price data, providing rich insight into aisle dynamics.
- Integration and Harmonization:After assembling the multitude of data assets, IRI identified key dimensions that needed to be aligned across the database such as time, geography, product, etc. in order to ensure a seamless integration. The data was then processed, sanitized, harmonized and stored at the “leaf level” (greatest possible granularity) to allow for complete accuracy and lightning fast access.
- Liquid Data Deployment:Following sanitation and harmonization, the complete database was then loaded into IRI’s Liquid Data™ platform. To make the data accessible and functional, IRI constructed brand hierarchies, assigned specific product attributes, distributed between specific segments, and developed all necessary calculations and rule sets.
Once the process was complete, IRI developed a unique analytic approach with the client to create a store-level assortment optimization tool. The solution optimized the potential for each store across a variety of execution tactics. The tool was based on store-level shopper preferences (loyalty to 25 key beer attributes) as well as constraints such as floor and shelf space. It was also dynamic, allowing the client to make decisions quickly for each of its 500,000+ stores. The resulting assortment recommendations were also delivered via mobile to each route driver for execution.