Method, sources and downloads
The data behind this site
Almost every figure here is counted straight from the Energy Performance of Buildings Register, the public record of every EPC lodged in England and Wales. One is calculated rather than counted, and it is labelled as such on its own page. Nothing is forecast, fitted or estimated.
For journalists
The tables and CSVs are free to reuse, including commercially, under the Open Government Licence v3.0. No permission needed, no paywall, no form. A link back is appreciated but not required.
If you want a cut we have not published, a figure for a specific constituency, or a sanity check on a number before you run it, ask. We would rather help you get it right than see it misquoted.
Email press@epcdeadline.com. A real person reads it.
Downloads
- Councils by volume below EPC C CSV. The number of privately rented homes rated below EPC C in each council area.
- Councils by share below EPC C CSV. The share of each council’s privately rented stock rated below EPC C.
- Postcode districts by volume below EPC C CSV. The number of privately rented homes rated below EPC C in each postcode district.
- Councils improving fastest CSV. Of the privately rented homes that were below EPC C and have since been assessed again, the share that reached C.
- Hardest rental stock to fix CSV. The share of sub-C rentals that assessors say could reach C on their recommended measures.
How the figures are counted
One row per property, not per certificate
A property can have several EPCs over the years. We keep only the most recent one for each, identified by its UPRN where it has one. Counting lodgements instead would overstate everything by about half.
Tenure comes from the assessor
"Privately rented" is the tenure the assessor recorded at the time of the inspection. It is the only tenure signal on the register, and it is not updated when a property changes hands, so treat it as a good indicator rather than a legal record.
England and Wales only
The 2030 rules and the Boiler Upgrade Scheme both stop at the border. Scotland and Northern Ireland run separate schemes and are excluded entirely.
One figure is calculated, and it says so
Everything on this site is a count from certificates, with a single exception: the like-for-like comparison, which standardises each council against its own housing stock so that building age stops doing the ranking. It lives on its own page, is labelled as derived at the top of it, and sets out its method and its limits in full. Nothing else on the site is modelled, estimated or projected.
What this is not
These counts cover property with a current certificate and a recorded tenure. That is the evidenced private rented sector, not the whole of it: a rental that has never needed an EPC will not appear. Published estimates of total rented stock are higher, and both figures can be right at once.
The numbers, as they stand
| Properties with a current certificate | 20,306,636 |
|---|---|
| Privately rented | 3,487,168 |
| Privately rented below EPC C | 1,846,062 |
| Share of the rented sector below EPC C | 52.9% |
| Heated by oil or LPG | 814,824 |
Policy figures on this site are stored with the date they were last checked against the source, and the site refuses to build with anything older than 90 days. EPC C deadline was last checked on 21 September 2026.
Licence and attribution
Contains public sector information licensed under the Open Government Licence v3.0. © Crown copyright. The source data is published by the Ministry of Housing, Communities and Local Government at get-energy-performance-data.communities.gov.uk.