A large electric utility has recently obtained regulatory approval to offer a Wind Power program to its residential electric customers. While the Utility already has a certain amount of wind resources in its overall generation portfolio, this new program will allow customers to purchase specific quantities of Wind Power in 100 kWh blocks, up to 100% of their entire monthly bill. The premium for this program is about $0.01/kwh, or about 10%. The Utility wants to understand which customers have a higher propensity to make purchases under this program, for purposes of improving the effectiveness of its marketing programs and program enrollment..
After obtaining customer demographic data and historical campaign responsiveness reports, Omni Analytics augmented these datasets with geolocation data to create a 360 degree view of the utility company’s consumer base.
Leveraging the historical campaign data as response variables, Omni Analytics used distance based supervised learning techniques to create a demographically identical look-a-like customer population for advance modeling.
The analysis provided clear attributes that correlated directly with enrollment propensity, directly supporting the business’s initial hypothesis that customers with higher incomes and greater interest in clean energy would have a higher propensity.
Based on this analysis, the Utility was able to focus its marketing efforts on those customers with the highest propensity for enrollment. Net result is increased marketing effectiveness and higher overall program enrollment.
Distance based Matching, Supervised learning, Look-a-Like modeling
Comments are closed.