AI—and more specifically, machine learning (ML) capabilities—has been a core part of Uplight’s product strategy since the company’s launch in 2020. Since then, we’ve continued to improve our data quality, ML models, and AI strategy to reach, motivate, and activate even more energy customers at scale.
Uplight’s data platform aggregates data from 22 million meters, 300 billion AMI data points, 50 billion telemetry data points, and 1.4 billion customer data points to create meaningful customer insights. Our models can:
- Predict customer propensity to install distributed energy devices (DERs) like EV chargers and water heaters
- Identify households with DERs
- Generate bill insights including usage disaggregation, month-over-month comparison, peer comparison, and forecast, surfaced in customer portals and eHERs
- Segment customers for hyper-targeted outreach
- Optimize DR performance and customer comfort by predicting load shed 24 hours in advance with 97% accuracy
Scaling participation and reaching 94% customer satisfaction with personalized experiences
Our models work together to scale customer participation across all Uplight solutions from Home Energy Reports (HERs) to VPPs. For example, EV detection has driven 8x more clicks from customers identified through these models compared to standard outreach. For DR programs, Predictive Capacity Dispatch (PCD) helps lower opt-outs due to shorter, more comfortable events and enables utilities to customize event lengths based on customer behavior and preferences.
Embedding personalization in customer touchpoints doesn’t only activate customers—it keeps them satisfied. With personalized insights and offers surfaced in HERs, we see customer satisfaction rates as high as 94%.
A sophisticated data platform with the help of Databricks
All of these models are dependent upon data—and lots of it. Uplight’s data platform runs on Databricks’ Lakehouse architecture, which unifies utility, weather, marketplace, and demographic data into a single, governed, discoverable location. This consolidation lets any Uplight team pull from one source of truth instead of maintaining duplicate pipelines, while powering all models across our products.
Databricks’ newly announced Genie One builds on our data platform, giving Uplight teams a conversational AI coworker that can query governed data directly, grounding answers in a live map of what the data actually means. Uplight is piloting Genie One to help internal teams understand and evaluate utility data faster—boosting efficiency, problem-solving speed, and innovation.
Micaela Christopher, Director of Data Science and Engineering, Uplight, writes: “Our investment in building the Uplight Data Platform on top of Databricks is paying off in powerful new ways. By bringing Genie One capabilities to our data, we’re enabling teams across Uplight to explore, discover, and innovate with more speed, confidence, and creativity than ever before. This is the promise of data democratization—enabling a culture where curiosity, data-informed decision-making, and innovation can happen at every level of the company.”
As Uplight continues to expand its AI and advanced modeling capabilities, the focus stays consistent: best-in-class data quality, customer experiences, and event performance for scaling utility Demand Stacks, while maintaining strict security and compliance standards
