How to Read UK Energy Statistics: Revisions, Samples and Official Status

Every few months a new energy statistic lands: how many homes have a heat pump, what share of households are struggling with bills, how much gas the country burned last year. The numbers look solid, but they arrive with labels, caveats and histories that decide how much weight they can bear. A figure from an accredited official statistics release is a different kind of object from a figure labelled "experimental", and a percentage of "households" can mean very different things depending on who was counted.

This page explains how to read those signals. It covers what the official designations mean, which surveys sit behind the housing numbers a householder actually meets, how samples are drawn, why figures get revised or stop being comparable, and how to tell a modelled number from a measured one. The English Housing Survey, for example, is an annual survey of 13,000 face-to-face interviews and 6,000 physical surveys of households in England¹, while the Scottish Household Survey invites a random sample of addresses across Scotland every year². Knowing that difference is the first step to using either set of figures well.

The meaning of "accredited" on a release

When a statistics release carries the label "Accredited Official Statistics", it has been assessed against the standards set out in the official statistics code of practice. The National Energy Efficiency Data-Framework (NEED) explains what that means in practice:

The label is not a claim about the numbers being final; it is a claim about the process that produced them.

Most of the headline energy releases carry it. Energy Consumption in the UK 2025 is published as accredited official statistics⁷, as is UK Energy in Brief 2026⁸ and the UK local authority and regional greenhouse gas emissions statistics, 2005 to 2024⁹. The English Housing Survey is an Accredited Official statistic, previously known as a National statistic¹⁰, and its energy reports and low carbon technologies fact sheets carry the same designation¹¹.

Below that sits a second tier. The Energy Performance of Buildings Certificates releases for England and Wales are published as Official Statistics¹³, and for several years were labelled "Official Statistics in development (formerly known as Experimental Official Statistics)"¹⁴. The energy efficiency characteristics of new dwellings are likewise published as Official Statistics in development, and the release states plainly that "These statistics are new and may be subject to refinement in light of user feedback"¹⁵. An earlier EPC release, covering July to December 2020, was published as experimental statistics¹⁶.

The practical rule is this: an accredited release has been through external assessment and its figures can be quoted with confidence in the process. A release in development can still be useful, and the new dwellings statistics are stated to be consistent with the Energy Performance of Buildings Certificates statistics release¹⁵, but its figures may move as methods settle. Neither label tells you the number is right for your question; both tell you how much scrutiny sits behind it.

Key surveys householders encounter

Two government surveys dominate the housing side of home energy statistics, one for each of the two largest nations.

The English Housing Survey is "a national survey of people's housing circumstances and the condition and energy efficiency of housing in England", a repeated cross-sectional survey that pairs a household interview with a physical inspection of the home¹⁰. It is one of the longest standing government surveys, first run in 1967¹⁰. Its energy reports give an overview of the energy efficiency of the English housing stock over a decade and then break the picture down by dwelling and household characteristics, and also report on the presence of different heating systems, whether households had changed their fuel supplier or tariffs, and the methods of payment households use to pay for their energy¹¹. A 2023 to 2024 fact sheet examines the prevalence and preparedness of English homes for low carbon technologies such as heat pumps and electric vehicle charge points¹².

The Scottish Household Survey provides "robust evidence on the characteristics, attitudes and behaviour of private households and individuals, as well as on the physical condition of Scotland's homes"². Its topics include housing, neighbourhoods, finances, internet, physical activity, local services, environment, volunteering, culture and childcare², so energy questions sit inside a much broader household survey.

A third survey appears in consumer-facing energy coverage. The Energy Consumer Satisfaction Survey is "a survey of domestic energy consumers in England, Scotland and Wales (Great Britain) commissioned by Ofgem and Citizen's Advice"³, covering domestic energy consumers across Great Britain¹⁷. It measures satisfaction rather than housing conditions, but it is produced to the same official standards and is often quoted alongside the housing surveys.

Around these sit the physical stock itself and the policy analysis built on it. Scotland has around 2.45 million homes, "all varied in type, use, size, age, construction, and energy efficiency"⁴, and more than 2 million of them use mains gas or oil as their primary heating fuel¹⁹. The Ministry of Housing, Communities and Local Government commissioned the Building Research Establishment to analyse English Housing Survey physical survey data from 2019 to estimate whether dwellings would pass or fail an updated Decent Homes Standard and the potential cost of bringing homes up to it²⁰.

How far sample sizes carry the findings

Every survey figure rests on a sample, and the size and design of that sample decide what the figure can honestly tell you.

The English Housing Survey is an annual survey of 13,000 face-to-face interviews and 6,000 physical surveys of households in England¹. The interview and the physical inspection are separate exercises: the interview captures the household's circumstances, while the physical survey, carried out by trained surveyors, captures the condition and energy efficiency of the dwelling itself. That is why energy efficiency findings from the survey carry more detail than most other sources, and also why they arrive with a lag.

The Scottish Household Survey works differently: "Every year, a random sample of addresses in Scotland are invited to take part in the Scottish Household Survey"². Sampling addresses rather than people is a deliberate design choice, and the survey explains its purpose: "This is to make sure the results are representative of everyone in Scotland, not just certain groups of people"². Against a stock of around 2.45 million homes⁴, the achieved sample is a small fraction, which is why results are published as estimates with confidence intervals rather than as counts.

The Ofgem and Citizens Advice satisfaction survey is smaller again, and sampled on a different basis: "It uses a quota-based approach to sampling"¹⁸. In the January 2025 wave, 3,854 domestic energy consumers took part³; an earlier wave, fieldwork conducted between 30 August and 18 September 2023, had 3,742 domestic energy consumers²¹. A quota-based sample is designed to hit set numbers within groups, which is a legitimate method but a different one from random address sampling, and the two should not be described as interchangeable.

What a sample supports is proportion, not count. A survey of 13,000 households can support a statement like "around two-thirds of Council or Housing Association houses have the highest EPC ratings A-C, compared to around 40% of houses in other tenure types"⁶, because that is a ratio estimated from a designed sample. It cannot tell you the exact number of houses at any rating; for that, you need the administrative datasets discussed below.

Breaks, revisions and discontinued series

Figures move after publication for ordinary reasons, and some series stop being comparable without any announcement being made. Reading a trend requires knowing where the joins are. There are five kinds to watch for:

• Methodological breaks. The clearest recent example is the Scottish Household Survey: "Due to limitations imposed by the Covid pandemic, the 2020 and 2021 surveys were carried out over the telephone. The results were published as experimental statistics and are not directly comparable with other years"²². Anyone charting Scottish household trends across 2020 and 2021 is working across a break in the series, and the two years should be treated as a separate block rather than part of the continuous line.
• Seasonal patterns. The EPC statistics carry an explicit warning about how they may be compared: the data "have a seasonal pattern and should only be compared between the same Quarters across years", and are not seasonally adjusted¹⁴. Comparing a spring quarter with the following winter quarter mixes a seasonal effect into the trend.
• Discontinued series. The ONS announced in April 2026: "we have reviewed our outputs and have decided to discontinue these Energy Efficiency of Housing statistics"²¹. A discontinued series does not become false, but it stops being maintained, and later releases from other producers will not line up with it.
• Regulatory changes. The Energy Performance of Buildings (Certificates and Inspections) regulations of 2007, 2008, 2009 and 2012 were all revoked in their entirety by the Energy Performance of Buildings (England and Wales) Regulations 2012²³. A certificate issued under one regime is not automatically equivalent to one issued under another, and long EPC time series cross these boundaries.
• Ordinary revision. Figures are revised as data comes in. The new dwellings energy efficiency statistics are explicitly subject to refinement¹⁵, and consumer tracking measures shift over time: satisfaction measures in the Ofgem survey are "now well below the tracking peaks recorded in 2019 and 2020"²¹, a comparison that only works because the methodology was held steady across those years. On the consumption side, NEED reports that the equivalent reductions for Scotland were 13% for gas and 12% for electricity, measured against 2021⁶, figures that depend on which base year the release chose.

Modelled, measured and administrative: three distinct figures

For any given question about homes, there are usually three kinds of answer on offer, and they are not the same kind of object.

• A measured figure comes from recording what actually happened: meters read, consumption summed. The Digest of United Kingdom Energy Statistics, the annual energy statistics publication, is built on this kind of accounting²⁴.
• A modelled figure comes from applying assumptions to data: the Building Research Establishment analysed English Housing Survey physical survey data from 2019 to estimate whether dwellings would pass or fail an updated Decent Homes Standard and the potential cost of bringing homes up to the revised standard²⁰. Those costs are estimates of what compliance would cost, not records of spending.
• An administrative figure comes from counting records that exist because of some process: certificates lodged, measures installed under a scheme, installations certified. The National Energy Efficiency Data-Framework's impact of measures data tables are accredited official statistics built by linking such records to consumption data²⁵. Ofgem's ECO reporting is administrative in the same way: monthly reports show "how energy suppliers are progressing towards reaching their targets for this obligation, and the number and types of measures installed"²⁶.

Some releases sit deliberately between categories. The pandemic-era rounds of the Scottish Household Survey were carried out over the telephone because of Covid limitations and published as experimental statistics, not directly comparable with other years. The label "experimental" was the honest signal of that status.

The reason the distinction matters is that the three kinds answer different questions. A modelled figure tells you what would happen under stated assumptions; a measured figure tells you what did happen; an administrative figure tells you what was recorded, which is not the same as what occurred, because records only exist for activity that had to be registered. When a heat pump share is quoted from EPC records, it covers homes with certificates, not all homes.

Understanding the denominator behind a percentage

The most common reading error is treating every percentage as a percentage of the same thing. The denominator decides what the figure means, and releases differ in what they count.

Take fuel poverty pressure. A House of Commons committee report states that more than a third of households in England were spending over 10% of their income after housing costs on home energy in 2023, and gives the figure precisely: "36.4% (8.91m) in 2023, up from 20.5% in 2021 (4.93m)"⁵. The percentage and the household count are the same fact expressed two ways, and the jump between 2021 and 2023 is a real change in household circumstances, not a change of method.

Now take heat pumps, where the denominator changes the answer three times over:

None of these contradicts the others; each is a share of a different group, and the group is stated in the release.

Attitude percentages carry the same need for care, because each is a share of the group actually surveyed. Tenure comparisons work the same way: in the DESNZ Public Attitudes Tracker for Summer 2025, 48% of people living in owner-occupied homes knew their home heating would need to change for Net Zero, versus 32% in rented accommodation, with private renters at 40% and social renters lower at 27%²⁷. Each figure is a share of its tenure group.

The habit worth building is simple: before quoting any percentage, find the sentence in the release that says "of what". If it is not stated, the figure cannot safely be used.

Publication sources and supporting documents

Behind every release sits a set of documents that explain how the figures were made, and knowing where they live turns a headline number into a usable one.

For the Scottish Household Survey, the publications collection holds the questionnaires from 2003 to 2019, annual methodology and fieldwork outcomes reports, and a privacy notice². Results are published on a Data Explorer divided into 11 topics: housing, neighbourhoods, household finances, internet, physical activity, local services, environment, volunteering, culture, childcare, and trust in public institutions²². The team can be contacted at shs@gov.scot².

For the English Housing Survey, the live tables are "updated each year and accompany the annual reports"¹¹, and the chapter structure of each report separates the headline findings from the detail on energy¹⁰. The survey's own description of its method, the questions asked and the physical survey process is the reference for anyone checking what a figure covers.

For the wider energy statistics, the Digest of United Kingdom Energy Statistics is the annual energy statistics publication²⁴. The local authority greenhouse gas statistics offer an interactive element: "Users can zoom in to any UK local authority, see the emissions for the area and identify the significant point sources", with data filterable by sector and across the time series⁹. The main data sources for that release are the UK National Atmospheric Emissions Inventory and the DESNZ Official Statistics on energy consumption for local authority areas⁹.

Scheme-level administrative data has its own home. Ofgem's ECO guidance page links to monthly public reports showing supplier progress and the number and types of measures installed²⁶. For anyone comparing releases, the methodology report is the document that settles whether two figures describe the same thing.

Locating your home within the figures

The point of reading all this carefully is to work out where your own house stands, and the surveys give several hooks for that.

Tenure is one. NEED's analysis found that around two-thirds of Council or Housing Association houses have the highest EPC ratings A-C, compared to around 40% of houses in other tenure types⁶. Social housing is rated better than the rest of the stock, so a household in a social home is, on the averages, in a more efficient dwelling than the national figure suggests.

Heating system is another. More than 2 million homes in Scotland use mains gas or oil as their primary heating fuel¹⁹, so the Scottish averages are dominated by fossil-fuelled homes. On heat pumps, the spread by property type is wide: 4.2% of new dwellings in England have heat pumps against 0.8% of existing dwellings, and detached properties lead at 3.2%²¹. A household in a detached home considering a heat pump is in the group where uptake is already highest.

Policy knowledge also varies by tenure, which matters for what a household can expect to be asked to do. The Summer 2025 Public Attitudes Tracker found 48% of owner-occupiers knew their heating would need to change for Net Zero, against 32% of renters overall, 40% of private renters and 27% of social renters²⁷. For privately rented homes in England and Wales, a consultation proposed raising the minimum energy efficiency standard to the equivalent of EPC C by 2030, raising the cost cap to £10,000 with a 10-year validity period for exemptions²⁸.

The physical readiness of the stock is better than the installation figures suggest. Around 90% of UK homes have sufficient insulation and electrical capacity to operate a heat pump, despite only around 1% having one installed⁵. The Climate Change Committee's assessment is that approximately 10% of existing homes in the UK will need to be heated by a heat pump by 2030, compared to only approximately 1% today²⁹.

Implications for a household's independence

Statistics about the housing stock are, in the end, statistics about dependence. They show how many homes rely on the gas network, how many could switch, and how fast the switch is actually happening.

The consumption figures show the dependence clearly. More than 2 million Scottish homes heat with mains gas or oil¹⁹, and NEED records equivalent reductions for Scotland of 13% for gas and 12% for electricity against 2021⁶, so the direction of travel is downward but from a high base.

The readiness figures show the headroom. With around 90% of UK homes having sufficient insulation and electrical capacity to operate a heat pump⁵, the physical barrier to independence from the gas grid is lower than the 1% installation rate implies. The gap between readiness and uptake is where policy, cost and information sit, and the tenure-split knowledge figures²⁷ suggest that many households do not yet know a change is coming.

For a household asking what the numbers mean for them, the honest reading is this: the stock is mostly heat-pump-ready but mostly gas-heated, the averages hide wide differences by tenure and property type, and every figure comes from a particular sample, model or register with its own limits. Reading the label, the denominator and the comparability statement is not pedantry; it is how you tell whether a headline number describes a home like yours.