This special interest group is where customer care professionals share tactics on how utilities are improving interactions with their customers. 

WARNING: SIGN-IN

You need to be a member of Energy Central to access some features and content. Please or register to continue.

Post

Advanced Analytics and Machine Learning in Utility Customer Management: A Matter of Life and Data

image credit: Wrightstudio - Dreamstime

This item is part of the Customer Care - Special Issue - 05/2019, click here for more

Two interesting phenomena are hitting the utility industry right now. One is the availability of deeper, richer data on customers, courtesy of smart meters. The other is the emergence of machine learning as a readily available tool to help improve all aspects of customer operations--from service order efficiency, to improved call center operations, to improved customer program engagement.  The intersection of these two major developments merits some exploration, but first let’s lay a bit of a foundation before we look at the reality of this in a large investor-owned utility.

First, there are some misconceptions about machine learning, so let’s start with a simple definition: Machine learning is a method of data analysis that automates analytical model building. It is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. This concept of learning, or training, from relevant data sets cannot be overemphasized. This is the core to what enables machine learning.

Machine learning is not a new concept. Early versions of machine learning have been around for decades, dating back to the “nearest neighbor” algorithm to maximize a salesman’s effectiveness to more recent examples like the infamous “Alpha Go” demonstration and self-driving cars.

Looking at applying machine learning in utilities, the customer management function is a great place to start, and tends to be a hotbed of opportunity for advanced analytic applications.  At SoCalGas, this is driven to a large degree by managing two aspects of the business:

  • Efficiency: Utilities want to run their operations as efficiently and effectively as possible. 
  • Enablers: The volume and velocity of data can now be captured and managed using technology that didn’t exist as recently as five years ago.  Also, the Data Scientist role is now becoming mainstream. These specialists have the skills and experience to more easily convert data to insights to action.

Like most utilities, SoCalGas is always looking for ways to continuously improve its business and manage operating expenses.  And within SoCalGas, the Customer Services organization is no exception.  Using advanced analytics is becoming infused in the company culture as a way to continuously increase efficiencies and improve customer service.  Several examples include:

  • Developing a weather elasticity model to identify those customers least likely to conserve during a winter cold snap, allowing SoCalGas to target them with relevant literature. 
  • Performing an analysis to determine the key drivers of bad debt, allowing SoCalGas to fine tune its methods of bill collection and reduce the number of customers sent to third-party collections agencies. 
  • Identifying customers most likely to contact SoCalGas to provide them tools and information to better manage their usage during the peak season and simplify the payment process to better enable customers to complete transactions in the channels of their choice (e.g., self-service).
  • Using advanced analytics to build a micro-targeting system that optimizes the paperless billing rate, whereas traditional marketing methods have limitations. 

To ensure SoCalGas achieves its customer experience goals, SoCalGas has an operating area entirely dedicated to measuring and tracking Customer Satisfaction through focus groups, surveys, and benchmarking research.

One by-product of these successful machine learning projects is that SoCalGas has seen a growing hunger for analytics to help solve business problems. For example, when SoCalGas began using machine learning a few years ago, its data scientists had to go hunting for problems they thought they could solve. Now they are starting to see operating areas coming to them and asking for help.

So, as the team at SoCalGas has demonstrated, advanced analytics and machine learning are not a futuristic technology that is going to take over the world, nor is it out of reach for utilities looking to leverage their rich data sources for business improvements across the enterprise. Watch for utility leaders to embrace these progressive and forward-thinking approaches as they strive to drive more value from their smart meters and other intelligent infrastructure investments.

Authored by By David Baron, SoCalGas, and Mike Smith, SAS Institute

David Baron works in Performance Mgmt & Org Strategy at SoCalGas. He is an operations and customer insight executive reflecting experience across multiple industries. David has a unique balance of executive business management, expert analytic skills, and translating data to value to action. He holds thee master’s degrees from DePaul University.

Mike Smith's picture

Thank Mike for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.

Discussions

No discussions yet. Start a discussion below.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »