Methodology

1. Source data

The analysis starts from the official England and Wales Energy Performance of Buildings bulk data, which publishes domestic certificate records and a separate set of recommendation records. We do not alter the source values: filtering decides which records are used, and the recorded values are carried through unchanged.

Full detail of the source material, its licence and its coverage limits is set out on the data sources page.

2. Current properties and UPRN deduplication

The analysis starts from 27,654,564 domestic certificate-history rows. Many properties have been assessed more than once, so certificates are grouped by valid UPRN and only the latest certificate for each property is kept. That leaves 19,324,312 current properties.

“Current” means the latest certificate recorded for each valid UPRN in this analysed release. A property that has been reassessed appears once, using its most recent certificate.

485,734 raw domestic certificate rows carry an invalid or null UPRN. Without a valid property identifier they cannot be reduced to one current property, so they do not enter the current-property comparison.

3. The variables a comparison uses

A comparison is only useful against homes that share the characteristics which drive an EPC score. This release aggregates on three recorded variables:

Postcode district

The first part of the postcode, such as M11. Postcode districts and sectors are attached using the ONS UPRN lookup; 19,172,788 current properties join successfully (99.21588929013359%). A full postcode is never requested or published.

Property type

House, flat, bungalow, maisonette, park home or unknown, exactly as recorded by the assessor.

Construction age band

The recorded age band, which stands in for typical wall construction, glazing and original heating design. Overlapping legacy source bands are preserved and clearly labelled rather than merged.

4. Published views and suppression

Rather than build a bespoke cohort for each property, this release publishes fixed aggregate views: all homes in a postcode district, the same property type in that district, the same construction age band in that district, and the same property type and construction age band nationally.

This aggregate release does not combine postcode district, property type and construction age in one published cell. The four views below are shown separately and must not be read as a single jointly matched cohort. The national property type and construction age archetype is the closest combined attribute view; the district views add geographic context.

5. What each published figure means

Each view publishes the number of properties, the median and mean current score, the mean potential score, the mean potential gap and the share of properties in each band A to G. The median is the middle score of the group and is used as the headline because a handful of very low or very high scores would drag an average away from the typical home.

A comparison is stated as the difference between your score and that view’s median. No percentile, no range, no individual property and no address-level result is produced, because the published data does not support one.

6. How recommendations are handled

102,824,532 domestic recommendation rows are counted by national property type and construction age band. For each archetype the most frequently recorded recommendations are published with the number of properties carrying them.

Recommendation figures are associations with a certificate's total potential SAP gap, not item-specific causal uplifts. The average shown is the average total current-to-potential certificate gap for properties carrying that recommendation. It is not the point uplift of that measure, and it does not show cause and effect.

No cost estimates are shown, because this release does not contain validated current costs.

7. Figures for this release

These figures describe the analysed release for England and Wales domestic current properties. This is not live data and it does not update automatically. The source snapshot ends 31 July 2026 and this release was generated on 23 August 2026.

Domestic certificate-history rows

27,654,564

All domestic certificate-history rows in the source snapshot.

Current properties (valid UPRN)

19,324,312

One current certificate per valid UPRN after deduplication.

Joined to district and sector

19,172,788

A join success of 99.21588929013359%.

Domestic recommendation rows

102,824,532

Recorded recommendation rows summarised by national archetype.

Rows with invalid or null UPRN

485,734

Excluded from the current-property comparison.

Published postcode districts

2,292

Each district has its own published aggregate file.

Limitations

• An EPC is a modelled assessment under standard occupancy assumptions, not a measurement of the energy a household actually uses.
• Comparability depends entirely on the fields recorded by the assessor, including property type and construction age band.
• This release publishes aggregate views only. It does not combine district, property type and construction age in one cell, so the views must be read separately.
• Groups smaller than 5 properties are suppressed, so some views are unavailable in smaller districts.
• Recommendation figures are associations with the certificate's total potential gap, not item-specific causal uplifts.
• No costs or savings are shown, because this release does not contain validated current costs.
• Not live data: the snapshot ends 31 July 2026 and the release was generated 23 August 2026.
• This is an independent service, not an official register and not a professional survey.

Update policy

For the position on any individual property, always check the official certificate on Find an energy certificate (GOV.UK).