Case Study: Large US Public Utility achieves 70% predictive accuracy identifying Time-of-Use (TOU) plan switchers with Blueocean Market Intelligence

Developed model to identify utility customers at risk of seasonal defection from Time-of-Use (TOU) price plan

Large US Public Utility engaged Blueocean Market Intelligence to identify which residential customers on a Time-of-Use (TOU) price plan were likely to switch to alternate plans. The utility’s challenge was to create individual propensity scores and uncover the factors that predict seasonal defection from TOU using its customer database and Meter Data Management System.

Blueocean Market Intelligence built predictive models (C5.0, CHAID, QUEST and logistic regression) on two years of customer, billing and meter data, producing propensity scores and profiles of Switchers vs. Non‑Switchers. The commissioned model achieved roughly 70% accuracy in flagging likely switchers, identified Energy Use and historical on/off‑peak patterns as key drivers, and enabled targeted retention strategies based on the modeled insights.