Case Study: AusNet Services achieves predictive maintenance, 50% cost savings and 3x faster data processing with Databricks

Using data to power millions of Australian households

AusNet Services, an Australian energy and infrastructure company serving 1.5 million customers and managing billions of electricity, gas and network connection assets, faced fragmented legacy systems and rapidly growing data (an expected 65+ TB over four years). Siloed warehouses and unstable compute made it difficult for engineers, analysts and data scientists to get timely insights, run experiments, and predict asset failures for proactive maintenance.

AusNet migrated production workloads to the Databricks Lakehouse on Azure—using Delta Lake, Databricks SQL, ETL and machine learning—to centralize data, speed processing and enable predictive maintenance. Within months they saw 3x faster processing, about 50% platform cost savings and a 20–30% reduction in operational overhead, improved cross-team collaboration, and expanded predictive maintenance across their asset fleet to reduce risk and maintenance costs.

AusNet Services

Souvik Das

Senior Data Platform Lead