Warm Home Discount Energy Cost Predictor

Modelling energy costs to determine eligibility for Warm Home Discount scheme

  • Publicly sourced

The challenge

The Warm Home Discount (WHD) scheme provides a £150 rebate for low-income households. Previously, eligibility required households to apply to an energy supplier for rebates, which were allocated on a first-come, first-served basis. This manual application process limited the scheme's ability to reach all eligible households, especially considering the 27 million households across England and Wales.

The solution

The system uses a log-linear regression model to predict energy costs based on property characteristics such as property type, age, and floor area, alongside an imputation model to fill gaps in missing data. These predictions are matched with benefits data from the Department for Work and Pensions (DWP) to identify eligible households. A rigorous appeals process allows customers to challenge decisions if they believe they are eligible for the scheme.

The results

  • In 2023/24, over 90% of the total 3 million rebates were issued automatically.
  • The implementation of this solution has made the process fairer by eliminating the first-come, first-served system. As a result, vulnerable customers who may not have been aware of the support available are now proactively identified and included.

Details

Organisation name
Department for Energy Security and Net Zero (DESNZ)
Government body
UK Government
User group
Civil servants
Use case type
Specific
Type of technology
Machine Learning
Phase
Live
Impact
Better Customer experience

Links

Get in touch

Email ai-knowledge-hub@dsit.gov.uk to:

  • find out how to use or collaborate on this tool
  • review content for this tool
  • talk to us about your own tool

Content created: 11 May 2025 | Last updated: 18 May 2025