Because they are measuring different things, and none of them claims otherwise. Four organisations published four different UK rental inflation figures for mid-2026. The gap between the highest and the lowest is wider than the lowest figure itself — and once you know what each one counts, the disagreement stops looking like error and starts looking like arithmetic.
Every article about rents picks one of these numbers. Almost none says which one, or why it differs from the number in the article next to it.
The short answer: one of them measures the whole rented stock, and the rest measure only the lets being agreed right now. Those two things move at different times, and they are currently moving in ways that put the stock measure above the new-let measures — the reverse of what most explanations assume.
The longer answer is that four of the five sources below can only see their own customers. That limitation is not hidden, but it is rarely stated next to the figure it produces.
Zoopla: 2.1%. UK rental inflation for new lets, in its Rental Market Report published 11 June 2026, with an average new-let rent of £1,321. Zoopla’s report →
ONS: 3.3%. The Price Index of Private Rents, for the 12 months to June 2026, average UK monthly rent £1,388, published in the ONS bulletin of July 2026. The ONS bulletin →
HomeLet: 4.3%. Its UK annual change for July 2026, average rent £1,369. Note that HomeLet also publishes an excluding-London figure, 3.4%, which is frequently quoted as though it were the UK headline — it is not. The HomeLet Rental Index →
Goodlord: 6.5%. For England only, June 2026, average £1,309 against £1,229 a year earlier. The Goodlord Rental Index →
A fifth, Rightmove, publishes a quarterly tracker of asking rents rather than a headline annual rate on the same cycle. Rightmove’s rental price tracker →
Zoopla describes its index as “a repeat transaction index, based on asking rents and adjusted to reflect achieved rents”. The input is its own portal listings; the index is produced by Hometrack, Zoopla’s data business. It covers new lets.
Rightmove has published a quarterly Rental Trends Tracker since 2011, built from the asking rents of properties advertised on its own site. Asking rents, own listings, new lets.
HomeLet works from its tenant referencing service, which it says covers over a million tenants a year through more than 5,000 letting agents, against a history of over 13 million tenancy applications. Its figures are “actual achieved rental prices with accurate tenancy start dates” rather than advertised costs, mix-adjusted with a methodology it says was designed with a professor at the London School of Economics. Newly agreed tenancies.
Goodlord reports on tenancies processed through its own platform, and covers England only.
ONS is the outlier, and not because of its statistics. Its Price Index of Private Rents is built from rental data collected by rent officers working for the Valuation Office Agency, the Welsh Government, the Scottish Government and the Northern Ireland Housing Executive — and they are not collecting it to produce an index. They collect it as part of a statutory duty to administer housing benefit, Universal Credit and Local Housing Allowance rates, under orders including the Rent Officers (Housing Benefit Functions) Order 1997 and the Rent Officers (Universal Credit Functions) Order 2013. The index is a reuse of administrative data gathered for paying benefit.
The ONS sample is a purposive one covering roughly a tenth of the UK private rental market, around half a million rents a year, mix-adjusted and modelled using hedonic double imputation. Crucially, it covers new lets, renewals and tenants sitting in an existing agreement.
This is the part that makes the other four numbers legible, and ONS states it directly in its own comparison of rental growth measures: “the greatest price rises occur when properties are newly let, compared with existing tenants, who tend to see smaller price increases”. The ONS comparison article →
Three honest qualifications on that quotation, because it is usually repeated without any of them. It was published in 2018. It describes IPHRP, the predecessor of today’s ONS index. And the evidence ONS offers for it is a landlord survey, in which the average increase on new lets was 5.0% against 3.4% on renewals. It is an ONS statement about a general tendency, not a rent officer’s observation and not a live measurement.
The consequence is structural. A measure built only from lets being agreed this month sees the sharp end of the market immediately. A measure built from the entire rented stock sees the same movement diluted across millions of tenancies whose rent has not been reviewed — so it registers later, and more gently. ONS puts the same point in its own terms: the index “measures the price change of the entire privately rented stock”.
Neither is wrong. They answer different questions. One tells you what it costs to move today; the other tells you what the country is actually paying.
The usual telling of this has the new-let indices racing ahead while the official measure trails behind. That was the shape of it while new-let rents were accelerating. It is not the shape of it today.
Today the ONS figure of 3.3% sits above Zoopla’s 2.1%. New-let growth has cooled, and the stock is still absorbing increases agreed a year or two ago, when it had not yet caught up. The gap has not closed; it has changed sign.
This is not evidence against the mechanism. It is what the mechanism predicts once the leading edge slows: the body keeps moving for a while afterwards. But it does mean the familiar sentence — “the official figures always understate what renters are paying” — is currently the wrong way round, and anyone repeating it is describing a market that has moved on.
Stock-versus-flow accounts for the official index sitting apart from the rest. It does nothing at all to explain the widest gap on this page.
Zoopla and Goodlord both measure newly agreed lets. Zoopla reports 2.1% and Goodlord reports 6.5% — three times the rate, for the same kind of transaction, over broadly the same period. No lag explains that, because there is no lag between them.
What explains it is who is in the sample. One is drawn from portal listings across the UK; the other from tenancies processed on a lettings platform in England. Different properties, different agents, different regions, different price points. The composition of the sample is doing the work, and composition effects can easily be larger than the trend being measured.
So there are two separate mechanisms on this page, not one. Stock against flow explains why the official measure is different in kind. Sample composition explains why measures of the same kind still disagree with each other.
Four of the five sources on this page observe their own commercial flow. Zoopla and Rightmove see properties advertised on their own portals. HomeLet sees tenants who pass through its referencing service. Goodlord sees tenancies processed on its own platform. Each is a sample of a customer base, and a customer base is not a market.
That is not a criticism, and none of them claims otherwise in their own methodology. It is simply a limit that travels with the figure and usually gets left behind when the figure is quoted.
ONS is the only one attempting to represent the population rather than a clientele — and even that is a purposive sample of roughly a tenth of the stock, gathered by rent officers doing a housing-benefit job, then reused for statistics. It is the most representative measure available and it is still a sample assembled for another purpose entirely.
There is no index of every rent paid in Britain. Every number in circulation, including the official one, is a partial view with a known shape. Knowing which shape you are holding is most of what it takes to read them well.
Which rental index should I use? It depends on the question. To know what it costs to rent a property now, use a new-let measure such as Zoopla, HomeLet or Goodlord. To know what is happening to rents across the country as a whole, including tenants who have not moved, use the ONS Price Index of Private Rents. They are not competing estimates of one quantity.
Why is the official ONS figure sometimes lower than the others? Because ONS measures the entire privately rented stock, including tenancies whose rent has not been reviewed, while the commercial indices measure only lets being agreed now. When new-let rents are rising quickly, the stock measure lags behind them.
So the official figure understates what renters pay? Not currently. In mid-2026 the ONS figure of 3.3% sat above Zoopla’s 2.1%, because new-let growth has cooled while the stock is still absorbing earlier increases. The relationship reverses depending on where the market is in its cycle.
Where does the ONS rental data come from? From rent officers working for the Valuation Office Agency, the Welsh Government, the Scottish Government and the Northern Ireland Housing Executive, collecting rents as part of their statutory duty to set Local Housing Allowance and administer housing benefit and Universal Credit. ONS reuses that administrative data.
Why do Zoopla and Goodlord disagree when both measure new lets? Because their samples differ. Zoopla draws on portal listings across the UK; Goodlord reports tenancies processed through its own platform, in England only. Different properties and regions produce different averages regardless of the underlying trend.
Is any index a complete picture of the UK rental market? No. Four of the five here see only their own customers, and the ONS index is a purposive sample of roughly a tenth of the market collected for housing-benefit administration. No source observes every rent paid.
UK Property Portal is written and published by Dan Woodcock, working independently from UK. It is a one-person operation — there is no newsroom and no research team behind it — and every figure on this page is attributed in the sentence that carries it, dated to the period it describes, and linked to the organisation that published it. Read the full editorial standard →
Rents for one local authority, not one headline for the country.
Average private rent by local authority and bedroom count, from the official ONS index — sourced, dated and free to look up.
Open the rent guides →Free. No account needed.