Ofgem Consumer Archetypes and Energy Segmentation Data

The Ofgem consumer archetypes are a set of statistical household groupings used to describe how energy is used, paid for and thought about across Great Britain. The 2024 update sets out 24 archetypes, split into ten groups ranging from A to J, and records that the total number increased from 13 to 24¹. Group sizes vary between 163,000 and 3.5 million households¹.

They are not a tool that tells a household which group it belongs to. They are clusters built from survey data, used to model how different kinds of home respond to tariffs, technology and policy. The 2024 update derived them from the Living Costs and Food Survey 2019-2020 and segmented them using a cluster analysis¹. That survey contained 5,050 households used to produce the archetypes¹.

For a household thinking about its own energy independence, the archetypes matter because they are the evidence base behind decisions about price controls, consumer protection and low-carbon technology support. They describe the population in segments, and the segments carry different levels of exposure to the grid, to gas and to the suppliers that serve them.

What the Ofgem consumer archetypes are

An archetype is a modelled household type. It bundles together the characteristics that tend to travel together in the data: income band, heating fuel, property age and type, tenure, age of the head of household, and how engaged the household is with the energy market. The 2024 update describes the result as a set of groups that can be used to understand consumption and behaviour across the population rather than case by case¹.

The purpose is analytical. Ofgem's principal objective is to protect the interests of existing and future consumers, including consumers' interests in the net zero 2050 target and five-year carbon budgets, and the security of supply of gas and electricity³. Archetypes give that work a population structure: which households are likely to be affected by a change, how many of them there are, and what their starting point looks like.

The segmentation also sits alongside a longer history of consumer categorisation in the sector. Earlier work divided consumers into behavioural groups rather than household types, including disengaged consumers at 20 to 30% of the total, permanently disengaged consumers at 20 to 30%, passive consumers at 20 to 30%, and increasingly sticky proactive consumers at 5 to 10%⁴. Those categories describe attitudes; the archetypes describe households. The two approaches answer different questions and are not interchangeable.

What the archetypes do not do is identify a property. A household cannot look itself up and be told it is archetype C9. The clusters are averages, and the survey base is a sample. The value is in the pattern, not the individual case.

From 13 archetypes to 24: how the segmentation was built and updated

The current structure is the product of a rebuild. The 2024 update records that the total number of archetypes increased from 13 to 24, and that the new set was derived from the Living Costs and Food Survey 2019-2020 and segmented using a cluster analysis¹. The earlier 13-archetype set was published in 2013 and used a different base, which is why the two generations are not directly comparable⁵.

The method matters for how the figures should be read. Cluster analysis groups households by similarity across many variables at once, so the boundaries between archetypes are drawn by the data rather than set in advance. That produces groups of very different sizes: the 2024 update gives a range of 163,000 to 3.5 million households per archetype¹. A large archetype is not more important than a small one; it is simply more common.

The update also separated off-gas households into their own cluster analysis, which is why nine archetypes are described as having no mains gas supply¹. That separation reflects a real difference in the options available. Off-gas households have limited opportunity to switch fuel source, so treating them as a variation on mains gas homes would flatten the distinction that matters most to them⁶.

The 2013 generation gives a sense of how the earlier segmentation was shaped. Archetype 8, younger working families in medium-sized rented houses, covered 10% of GB households and 2.5 million households; archetype 11, asset rich empty nesters in detached houses in less urban areas, covered 10% of GB households and 2.5 million households; archetype 7, low-income single adults in social rented houses, covered 5% of GB households and 1.2 million households⁵. The 2014 report behind that work gives counts of 2,580,000 for the asset-rich empty nester group and 937,000 for low-income out-of-work single adults in small one-bed social rented flats in London⁷.

The ten groups A to J: sizes, data sources and method

The 24 archetypes are split into ten groups ranging from A to J¹. The letter denotes the group; the number denotes the archetype within it. So A1, A2 and A3 sit in group A, and G17 and G18 sit in group G.

The consumer consent figures published in January 2026 give household counts and average annual electricity consumption for a number of named archetypes. These are the most granular published numbers for individual archetypes.

All figures in the table are from the consumer consent impact assessment². The spread is wide: average annual electricity runs from 2,742 kWh in A1 to 6,597 kWh in B5, and the presumed uptake of the consumer consent solution ranges from 0.5 to 2% in A1 and H19 up to 2 to 11% in B4². The expected bill reductions are modelled outcomes for each archetype, not offers available to households.

The data sources behind the segmentation are official surveys rather than commercial panels. The Living Costs and Food Survey 2019-2020 is the base for the 2024 archetypes, and it is an annual survey of around 5,000 households across England, Scotland and Wales¹. That is a sample of the population, not a census, and the archetypes inherit its limits.

Off-gas archetypes: which groups they cover and why it matters

Nine archetypes were produced from the off-gas cluster analysis: B4, B5, C8, F15, F16, G17, G18, H19 and J24¹. These are the archetypes without mains gas supply. They span several letter groups, which reflects the fact that being off the gas grid cuts across income, tenure and property type rather than sitting in one part of the population.

The reason they are separated is practical. Off-gas households have limited opportunity to switch fuel source, so the usual routes to changing what a home runs on are not open to them in the same way⁶. A household on mains gas can move between gas and electricity for heat, at a cost. A household on oil, LPG or solid fuel has fewer options and often faces higher and more volatile input costs.

The archetype data shows how varied that group is. H19 is described as upper-middle income, oil heating, rural, with a poor EPC, and covers 675,712 households². G17 is upper middle income, oil or other heating, unconventional housing, rural, and is the smallest archetype in those figures at 163,166 households². G18 is upper middle income, other heating fuel, rural and owner occupied, at 667,836 households². B4 is low income, electric heating, aged 65 and over, at 731,318 households². B5 is the archetype where 99% of households are on disability benefits¹.

An earlier off-gas archetype, H13, was described as off gas, low income, high electricity consumption, disability benefits, over 45s, low energy market engagement and late adopters, with 12% of households having children⁸. That combination, high consumption, low engagement and no gas connection, is the hardest version of the problem the segmentation is designed to expose.

Segment A, B and G: the three profiled segments in detail

Groups A, B and G carry the clearest published detail, and they illustrate how differently the segments behave.

Group A covers the lower-consumption archetypes. A1 has 578,333 households and average annual electricity of 2,742 kWh, the lowest in the published set, with presumed uptake of the consumer consent solution at 0.5 to 2% and an expected bill reduction of £35 to £50². A2 has 868,191 households at 2,849 kWh, with uptake of 5 to 8% and a reduction of £45 to £80². A3 has 883,413 households at 3,519 kWh, with uptake of 1 to 3% and a reduction of £40 to £65². The pattern is a large, relatively low-consuming population with modest modelled engagement.

Group B is where consumption and vulnerability rise. B4 has 731,318 households, average annual electricity of 4,811 kWh, presumed uptake of 2 to 11% and an expected bill reduction of £55 to £95². It is also one of the off-gas archetypes, described as low income with electric heating and a head of household aged 65 or over². B5 has 465,288 households and the highest average annual electricity in the published set at 6,597 kWh, with presumed uptake of 2 to 4% and an expected reduction of £40 to £60². In B5, 99% of households are on disability benefits¹.

Group G is the rural, off-gas, higher-cost segment. G17 has 163,166 households, average annual electricity of 5,901 kWh, presumed uptake of 2 to 5% and an expected bill reduction of £90 to £165, the highest lower bound in the published set². G18 has 667,836 households at 5,294 kWh, with presumed uptake of 2 to 5% and an expected reduction of £100 to £170². Both are off-gas archetypes, and both are described as upper middle income and rural².

Read together, the three groups show that the highest modelled bill reductions sit with the rural off-gas households, while the largest populations sit with the lower-consumption groups. That is the tension the segmentation is built to make visible.

What the archetypes say about heat pump and solar uptake

The archetype data itself does not publish heat pump or solar uptake rates by segment. What it does is identify which households are most likely to be affected by the transition, and the wider official statistics show how far that transition has to run.

Heat pumps remain a small share of heating. In the 2021 consumer survey, heat pumps were the main heating system for 1% of GB energy consumers, and take-up was 6% among 16 to 34 year olds⁹. Among those with heat pumps, 45% were on a green tariff, compared with 37% of those with solar PV and 35% of those with smart appliances⁹.

Awareness is much higher than adoption. In the summer 2025 public attitudes tracker, awareness of air source heat pumps was between 86% and 87% for people in age groups 55 and over, falling to 59% of those in younger groups, and 84% of people in owner-occupied households said they were aware of them¹⁰. Intention to install is far lower and tracks concern about climate change: 32% of those very concerned said they would be likely to install an air source heat pump, compared with 5% of those not concerned¹⁰.

On the consumption side, heat pump electricity use by other final consumers was 4.0% of total electricity demand in 2024, and 23% higher than the 2014 value¹¹. That is a measure of the whole category, not of domestic installations alone, but it shows the direction and the modest scale.

The archetypes connect to this through the off-gas groups. Households in B4, G17, G18 and H19 are the ones for whom a change of heating system is most consequential, because they have no mains gas to fall back on and, in the case of B4, low income and electric heating already¹. The segmentation does not predict what any of them will do. It identifies where the decision is hardest.

How Ofgem uses the archetypes: policy, price controls and the Consumer First Panel

The archetypes feed into policy work rather than standing alone as a research output. Ofgem's principal objective is to protect the interests of existing and future consumers, including their interests in the net zero 2050 target and five-year carbon budgets, and the security of supply of gas and electricity³. The 2026 review of Ofgem sets out a new set of three equal principal objectives focusing on the interests of existing and future consumers, net zero and growth¹².

The Consumer First Panel is the deliberative counterpart to the quantitative segmentation. It brings a broad group of domestic energy consumers together three to four times over the course of a year to feed into Ofgem's policy making¹³. Year 9 Wave 1 spoke to 75 consumers, and its first event focused on the needs of consumers in relation to information and communications from suppliers¹³. The panel's purpose is to understand what consumers need both to deal with their immediate energy needs and to make optimal decisions about which suppliers to use or switch to¹³.

Panel topics have included energy network price controls, the subject of Year 7 Wave 2¹⁴. Year 7 Wave 3 asked which hypothetical future business models appeal to consumers and why, and which business models might encourage consumers to engage more with the energy market¹⁵. The panel brings a broad group of domestic energy consumers together 3-4 times over the course of a year to feed into Ofgem's policy making, and one wave spoke to 75 consumers¹⁵. The panel's research has covered in-depth analysis of billing information and price metric information¹⁶.

The panel is qualitative and small by design. It does not assign households to archetypes, and its findings are not statistically representative in the way the segmentation is. What it adds is the reasoning behind consumer responses, which the clusters cannot supply.

The archetypes also sit within a wider consumer protection framework. Ofgem's market compliance review of how suppliers support customers in vulnerable situations found that all suppliers had issues to address, divided into three groups: minor weaknesses, moderate weaknesses and severe weaknesses¹⁷. The informed choices principles require suppliers to put in place information, services or tools to enable consumers to easily compare and select tariffs that suit their characteristics or preferences¹⁸. The government has also worked with Ofgem on the publication of a wider range of supplier performance data¹⁹.

Where the archetype data falls short for individual households

The limits are structural, and they are worth stating plainly because the archetypes are sometimes read as more precise than they are.

The base sample is small relative to the population. The Living Costs and Food Survey 2019-2020 contained 5,050 households used to produce the archetypes, drawn from an annual survey of around 5,000 households across England, Scotland and Wales¹. That is a robust sample for population-level analysis and a thin one for any single archetype, particularly the smaller groups.

Coverage is not uniform across the UK. The survey base covers England, Scotland and Wales¹. Northern Ireland is not part of that sampling frame, so the archetypes describe Great Britain rather than the United Kingdom, and figures for Northern Ireland households come from separate sources.

The archetypes are averages, not addresses. A household in a rural off-gas property with a poor EPC may resemble H19 in some respects and G17 in others, and the segmentation offers no way to resolve that. The clusters describe the shape of the population, not the position of a home within it.

There are also limits in the related data that sit alongside the archetypes. The customer credit balance figures published by Ofgem only relate to households that are in credit; including households in debit or with a balance of £0 would lower the average²⁰. That is a reminder that published averages are shaped by what they exclude, and the same caution applies to any figure built on a survey sample.

Finally, the archetypes are a snapshot. The 2024 set is built on 2019-2020 data, and the 2013 set on an earlier base, so comparisons across generations describe different populations as well as different methods¹. The Living Costs and Food Survey from 2019-2020 contained 5050 households that were used to produce the archetypes⁷. Where two figures for the same archetype differ, both should be read with their date attached.