Case Study: PepsiCo gains store-level predictive insights and frees 4,300 workdays with Microsoft Power BI

PepsiCo uses Azure Machine Learning to identify consumer shopping trends and produce store-level actionable insights

PepsiCo, a global food and beverage leader, needed to better match consumer demand with inventory across thousands of U.S. stores and move from manual reporting to predictive analytics. To do this the company deployed Microsoft technologies — including Azure, Azure Machine Learning, and Microsoft Power BI — to centralize data and pilot machine-learning models that produce store-level, actionable insights.

Using Azure Machine Learning’s MLOps capabilities and a Store DNA application, PepsiCo delivered daily, prioritized recommendations to field associates and executive dashboards in Microsoft Power BI. The pilot achieved 85% adoption of recommendations, improved predictions by more than 40%, scaled to 14 markets with 28 models tracked, and is estimated to free about 4,300 workdays a year from routine tasks to higher-value activities.

PepsiCo

Michael Cleavinger

Senior Director of Shopper Insights Data Science and Advanced Analytics