E Source OneInform’s data science puts predictive data to work for you

Feeling overwhelmed by all the data your utility gathers? Not sure what to do with it? You’re certainly not alone. That data has a lot of power if you just know how to unlock it. E Source OneInform holds the key. As a leading data science company, we use your valuable predictive data and put it to work for you, and we heard real-world utility success stories during our recent webinar Next-generation programs for an equitable clean-energy future.

How can OneInform change the way you look at data?

Tom Martin, E Source Data Science managing director, began by explaining how OneInform provides the next step in the evolution of customer engagement. Traditionally, utilities have tried to understand their customers by dividing them into large groups or segments. But, by harnessing the power of data science services and data analytics, OneInform gives utilities the ability to focus on specific customers, and then dynamically predict each customer’s fit for various programs the utility offers. “If you better understand your customers as unique, individual customers—a concept that we call an ‘audience of one’—then you can take action to engage your customers in a way that creates more value for them, and more value for you and your programs,” Martin said.

Data tells a story you can learn from

During the webinar, three guests explained the different ways they leveraged data science to better understand their customers and predict future customer behavior. In each case, OneInform provided tools to answer specific questions, empowering utilities to meet their customers’ needs while achieving their program goals.

Match the right customers with the right programs. Before collaborating with E Source, Portland General Electric (PGE) had been viewing its residential customer base through the lens of static macropersonas. Without a thorough understanding of its customers as individuals, however, the utility couldn’t predict which customers would be most successful and satisfied with its programs.

OneInform divided PGE’s customers into five distinct microsegments according to projected load shift and behavioral factors influencing satisfaction and outreach.

Kathy Wagner, PGE product manager of demand response, explained that the utility supplied one year of customer predictive data to E Source to create usage-based impact models.

Then E Source’s data scientists used OneInform to divide PGE’s customers into five distinct microsegments according to projected load shift and behavioral factors influencing satisfaction and outreach. The microsegmentation process showed that PGE’s individual customer impacts were wide-ranging. For example, only 14% of peak-time rebate enrollees accounted for 58% of the utility’s overall program load-reduction impact.

Thanks to the insights offered by OneInform’s microsegmentation of its customer base, PGE can now:

• Do a better job of predicting which individual customers are most likely to enroll in its programs
• Target outbound marketing efforts to reach specific customers who will be most successful and satisfied with each program
• Improve customer equity in programs and balance the enrollment mix
• Design new programs to optimize lifetime customer value
• Better forecast the local grid impact
• Support the utility’s decarbonization and energy-efficiency goals

Predict which customers are most likely to have trouble paying bills. When the Los Angeles Department of Water and Power (LADWP) wanted to study the financial impact of COVID-19 on its customers, the utility asked E Source to help. LADWP was specifically trying to identify customers who were likely to have a high risk of payment problems during the pandemic.

Erin Carone, senior utility services specialist at LADWP, explained that the utility had collected plenty of data, but didn’t know how to analyze it for its predictive capability. So LADWP started by providing E Source with a batch of historical data (including billing, payment, contact center, and arrears information). OneInform then combined that data with behavioral attributes and other unique customer information to create tools for the utility, including propensity models and clustering customers into predictive cohorts.

Thanks to the power of this new analysis, LADWP is now able to:

• Understand its customers better as individuals, including identifying which ones are struggling to pay bills or are likely to pay bills late
• Know the best ways to reach out to those customers
• Target communications to the right customers, offering them tailored payment programs

“We’re so overwhelmed with how much information is out there: Where do you start? What do you do? And that’s really where the value of E Source has come in.” —Jeremy Cappello, DTE

Plan programs that your customers really want and need. When DTE Energy was trying to understand how its small and midsize business (SMB) customers could take advantage of its energy waste reduction programs, the utility approached E Source for help.

First, the utility needed to know which programs would appeal to its SMB customers. Although DTE already had over 12 years of participation data and various segmentation models, the answers weren’t clear.

“The challenge that we have is, what the heck do you do with all this data?” said Jeremy Cappello, DTE’s energy waste reduction program lead—small, medium business. “We’re so overwhelmed with how much information is out there: Where do you start? What do you do? And that’s really where the value of E Source has come in.”

The utility asked E Source data science to help create a propensity score to understand what motivates its SMB customers, and how to engage them so that its programs could be more cost-effective.

Going forward, OneInform will help DTE understand its customers on an individual basis and make predictive data-driven decisions, including:

• Identifying the right customer to fit each program
• Finding the best time to engage customers when they’ll be receptive to a program offer
• Meeting customers where they are: knowing the right way to reach them and the most efficient way to market to them—through social media, emails, flyers, or other methods
• Planning and developing future programs with greater accuracy based on knowing its customers better

In the end, program success isn’t only about counting the number of enrollees. It’s about getting the right customers into the right programs. Have a question the data science services experts at E Source can answer? Send us a message and we’ll explain how OneInform can meet your utility’s needs.