Download PDF
C5i > Case Studies > Multivariate profiling to understand utility customers by level of engagement
C5i Logo

Multivariate profiling to understand utility customers by level of engagement

Technology Category
  • Analytics & Modeling - Big Data Analytics
  • Analytics & Modeling - Predictive Analytics
Applicable Industries
  • Utilities
Applicable Functions
  • Business Operation
  • Sales & Marketing
Services
  • Data Science Services
The Challenge
The client, a large US public utility, was looking to develop an engagement scoring system for profiling highly engaged residential customers as input to product/service marketing efforts. The challenge was to measure engagement based on the number and importance of interactions by the customer with the utility on four different dimensions. The goal was to establish a distinct segment of most engaged customers for targeted marketing of new product offers and ways to increase interaction with and convert less engaged customers.
About The Customer
The customer in this case study is a large public utility company based in the United States. The company operates in the utilities industry, providing essential services to residential customers. The company was looking to better understand its customers and their level of engagement with the company's services. The goal was to identify highly engaged customers and use this information to inform their product and service marketing efforts. The company was also interested in finding ways to increase interaction with less engaged customers and convert them into more engaged customers.
The Solution
Blueocean Market Intelligence developed a framework to measure engagement based on the number and importance of interactions by the customer with the utility on four different dimensions. Engagement scores were first developed for each dimension based on several parameters, and then an overall engagement score was computed using weights developed through an analytic hierarchy process. Third-party data was used to profile the groups of customers based on demographic, attitudinal and behavioral variables, as well as customer transaction data available from the client. Finally, a profile-based predictive model was built using a decision tree approach to predict which customers were likely to have a high propensity of engagement with the utility, based on demographic, attitudinal, behavioural and historical transactional data.
Operational Impact
  • Five groups of customers were identified based on their level of engagement with the utility.
  • Established top three rules to predict high engagement levels using existing customer data.
  • Identified characteristics of most engaged customers who are most likely to be open to new product offers.

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.