IrelandEnergy and utilities2025 – ongoingSSE Airtricity
Energy Advisor with personalized saving advice for smart meter customers
SSE Airtricity sent AI-written energy-saving advice to more than 65,000 smart meter customers, and visits to its insights page rose 12% in the first month
Problem
SSE Airtricity supplies energy to around 750,000 customers in Ireland, and many have smart meters that record 48 readings a day. The company wanted to turn that data into personal advice on cutting energy use. A competitor made customers wait 120 days after signup for insights.
Approach
The in-house data and analytics team built the Enhanced Smart Insights Energy Advisor on the Databricks platform. It uses Anthropic's Claude Sonnet 4.5 through Databricks model serving to write advice for more than 65,000 smart meter customers. AI judges check each insight for length, language, tone and accuracy, and insights that fail are regenerated. The advisor soft-launched in mid-December 2025, and insights refresh every two months when customers receive their bills.
Results
Takeaway
Automated quality checks on every generated message let the team send personalized advice at scale without reviewing each one by hand.
Categories
- Size
- Large enterprise
- Industry
- Energy and utilities
- Business function
- Customer service
- Innovation type
- Product and service innovation
- AI technology
- Generative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how SSE Airtricity built an energy advisor on Claude Sonnet 4.5 that gives personalized saving advice to more than 65,000 smart meter customers. The page reports a proof of concept to production time of about four months. It also reports a 12% rise in web visits to the insights section in the first month, and that 90% of logged-in customers went to the section and 10% viewed their full set of insights.
Full details are in the original sources.
