Bain uses cookies to improve functionality and performance of this site. More information can be found in our Privacy Policy. By continuing to browse this site, you consent to the use of cookies.

Case Study

Advanced Analytics powers up UtilityCo’s reliability, and customers notice

Despite UtilityCo’s* significant investment in infrastructure, the reliability of its service remained an issue. But that didn’t stop the company from setting aggressive goals for reliability even as it also vowed to cut costs. How to square that circle? UtilityCo partnered with Bain to harness the power of advanced analytics and data modeling in order to spot potential power outages before they occur, respond more effectively and boost customer satisfaction levels.

  • min read

At a Glance

  • ~3x Ability to predict equipment failures
  • 2x-3x Ability to forecast outage duration accurately

The Full Story

The Situation

Electricity outages can cause major disruption to energy customers, as well as to the provider’s brand and customer satisfaction ratings. UtilityCo is one of the largest electricity providers in the country, serving more than 14 million customers. For several years, the company made large capital investments to upgrade its vast power network, yet its reliability rankings remained stubbornly stuck below its top-performing peers.


UtilityCo kept key data points that could help improve reliability, such as customer characteristics, grid assets and operational performance, in silos across the organization. Even if the data were accessible, UtilityCo lacked the in-house talent and know-how to perform advanced analytics on complex data sets and apply those insights to its reliability efforts.

Our Approach

Bain began its work with UtilityCo by assessing the many factors that contribute to a power outage and then delineating the stages of an outage. Next, Bain helped UtilityCo restructure and analyze its data, with a focus on:

  • Using expanded data and criteria to predict the likelihood of equipment failure;
  • Determining the impact of equipment failure on high-risk customers to develop a “consequence score”;
  • Optimizing outage response; and
  • Managing customer expectations with accurate restoration predictions and customized communications.

Recommenations

UtilityCo and Bain applied advanced analytics to the company’s structured data set, yielding several key actions and recommendations.

Structure and expand the data. The Bain team merged more than 10 previously siloed data sets from various functions and sources, including Big Data from smart meters. Using predictive models, the Bain team enabled a two times to three times improvement in UtilityCo’s ability to identify transformers with the highest risk of failure. As the data set that powers that model expands, there is strong potential to achieve a four times improvement.

Create new consequence and risk models. While all outages need a prompt response, not all have the same consequences. Outages that affect hospitals, high-traffic areas or first responders can be particularly catastrophic. Bain helped UtilityCo integrate these factors into equipment repair prioritization to ensure that high-risk areas get a particularly fast response. UtilityCo now reviews the risk scores and failure risk when planning work to leverage opportunities for bundling.

Optimize outage responses. Prior to the development of advanced analytics tools, UtilityCo had limited visibility into the drivers of outage duration. The Bain team mined unstructured data and paired it with GPS locations to provide a full end-to-end view of the steps in the outage process. They then conducted statistical analysis to figure out what would drive higher durations (for instance, locations, weather and so on). This work enabled UtilityCo to deploy a field tool to track durations and also led to many changes in outage procedures.

Improve customer communication and satisfaction. With new predictive capabilities and enhanced customer data, UtilityCo can better manage customer expectations by providing personalized, accurate communications regarding power restoration. This will contribute to improved customer satisfaction, especially as measured by UtilityCo’s PQR (power, quality and reliability) score, a critical metric by which regulators evaluate utility companies and benchmark them against their peers.

 

The Results

With Bain’s help, UtilityCo used advanced analytics and predictive modeling to improve many operational factors that contribute to reliable service. Among the notable results:

  • Outage prevention. The development of an asset risk scoring model nearly tripled UtilityCo’s ability to identify its riskiest transformers, and it helped the organization gain a much more detailed understanding of relevant variables and impacts.
  • Optimized outage responses. Using data visualization and process changes derived from an analysis of the causes of outages, UtilityCo was able to reduce overall average outage duration by nearly 15% (90 minutes) in two years.
  • Improved customer service. UtilityCo has identified the potential to improve customer satisfaction PQR scores by up to 85 points, stemming from a two times to three times improvement in predicting restoration times combined with the ability to provide customized communications.

 

* We take our clients' confidentiality seriously. While we've changed their names, the results are real. 

Want to continue the conversation?

We help global leaders with their organization's most critical issues and opportunities. Together, we create enduring change and results.