Australian energy company, Jemena, has collaborated with professional services firm, Deloitte, to build a new cloud-based application on Amazon Web Services (AWS) that delivers valuable insights from its smart meter data to help more than 360,000 residential and business customers across Melbourne’s north-west make better energy decisions.

By leveraging the cloud-based application, Jemena can better understand customer consumption habits, which has allowed it to personalise its communications to help support Victorian electricity customers during the coronavirus (COVID-19) pandemic.

Commencing last year, Jemena and Deloitte, an AWS Partner Network (APN) Premier Partner, collaborated to build an application on AWS, leveraging machine learning services including Amazon SageMaker, a fully managed service that makes it easy to build, train, and deploy machine learning models quickly.

Jemena and Deloitte used Amazon SageMaker to develop algorithms and machine learning models, which are applied to smart meter data collected by Jemena to uncover commonalities and differences in how its customers consumed electricity.

“We looked at consumption behaviours such as how customers’ energy use changes during the day, peak usage hours, and how weather conditions such as extreme heat impact energy use for our residential and small business customers,” detailed Andrew Davis, Jemena General Manager for Electricity Strategy and Commercial.

These consumption insights helped Jemena identify patterns in energy usage, and in turn, provide more relevant communications and advice to their customers such as energy-saving tips.

With more people required to work from home during the on-going pandemic, many customers are unaware of the extent of the increase in their energy consumption.

Jemena is now leveraging the consumption insights generated by the application to deliver communications via SMS, direct mail, and email to tailor support to customers’ individual energy needs, assisting them to take more control of their energy usage, and avoid the potential for bill shock.

By using AWS services such as Amazon Athena to analyse large amounts of historical smart meter data, and Amazon SageMaker to gain more accurate customer insights into Jemena’s customer segments using machine learning, the energy company has also been able to identify at-risk customers who may not be digitally engaged, to ensure it can change its communications approach to direct mail so they are informed about their energy consumption.

“With years’ worth of smart meter data sitting in our data warehouses, we looked at how to migrate this information to the cloud and use it in an intelligent way to empower our electricity customers at a time that they need it most,” Mr Davis shared.

“Using these smart meter insights, we have been able to notify customers with significantly higher monthly consumption via SMS with a link to keep track of their energy usage through Jemena’s online customer platform, the Electricity Outlook Portal. Direct mail letters are sent to customers that have significantly decreased their energy usage with information about bill payment support and a free over-the-phone energy advice service.”

“Our customers have told us they want more control over their energy consumption and more personalised communications. Working with Deloitte and AWS has allowed us to deliver this commitment to our customers. By changing the way we analyse our smart meter data and leveraging advanced machine learning technology, not only have we opened up new opportunities to support customers during COVID-19, it will also help drive our long-term strategy to improve the customer experience,” Mr Davies commented.

According to Jahanzeb Azim, Deloitte Analytics and Cognitive Partner, Jemena had over 11 billion consumption data points from household smart meters to analyse by leveraging the scale of AWS – drawing a parallel to Netflix viewing time; that is equal to three million hours or 350 years’ worth.

“We were eager to collaborate with Jemena and leverage AWS to turn this significant smart meter data asset into information that could help them better understand their customers’ energy usage and drive better business decisions,” Mr Azim said.

Karl Durrance, AWS Director of Enterprise in Australia and New Zealand, added: “As we adapt to significant changes in how we live and work this year, many organisations are leveraging technology such as machine learning to respond to customers’ needs and drive product and service innovation.”

“We are excited to collaborate with our APN Premier Partner, Deloitte to help Jemena leverage the power of data to create personalised experiences and help Victorians to manage their energy consumption.”