Estate Agent & Property Marketing Statistics 2026: Portal Costs, Time to Sell and Lettings Economics

UK estate agency has one of the strangest marketing economics of any sector. Almost every buyer and tenant starts in the same place, on a portal the agent does not own, and the agent pays a monthly fee for the privilege of being there.

That fee is now the single largest fixed marketing line for most independent branches. Rightmove alone charged the average agency branch £1,636 a month in the first half of 2026, and it has raised that figure every year for a decade.

The numbers below come from Rightmove plc's own audited and unaudited results, Rightmove's House Price Index, Zoopla's House Price Index, Propertymark's monthly member survey, TwentyCi's Q2 2026 homemover data, Foxtons Group's annual report, HMRC transaction statistics and the ASA rulings database. Everything here was pulled from the primary source in August 2026.

Key estate agent and property marketing statistics for 2026

• 8.4 billion minutes were spent on Rightmove in the first half of 2026, down from 9.1 billion in H1 2025, according to Rightmove plc [1]
• Rightmove's share of time spent on UK property portals reached 90% on Comscore and 75% on SimilarWeb/Sensor Tower in June 2026, up from 80% and 74% a year earlier (Rightmove plc) [1]
• Less than 0.5% of Rightmove's traffic was referred from large language models in H1 2026, whilst over 85% was direct and organic (Rightmove plc) [1]
• Agency ARPA hit £1,636 per branch per month, up 8% year on year, which works out at roughly £19,600 a year per branch before Zoopla, OnTheMarket or anything else (Rightmove plc) [1]
• Rightmove ended June 2026 with 16,591 agency branches, up 1% on December, with agency retention at 96%, its highest first-half figure in over ten years (Rightmove plc) [1]
• Rightmove says its products delivered 7 out of 10 vendor instructions in resale and 8 out of 10 tenants for lettings properties in 2025, citing Street and RLTS survey data [2]
• The average home took 62 days to find a buyer in June 2026, with agents holding an average of 65 properties each, per the Rightmove House Price Index [3]
• Propertymark members recorded just 2.2 viewings per available property and 8.0 sales agreed per branch in May 2026, with 84% of agents reporting sales below asking price [5]
• The average newly listed UK property in Q2 2026 was priced 11.6% above its automated valuation, up from 5.7% a year earlier, according to TwentyCi [6]
• GetAgent puts the UK average estate agency fee at 1.43%, or about £3,966 on a £268,000 sale [9]
• Lettings generated 64% of Foxtons Group's revenue in 2025, against 30% from sales, and revenue per sales transaction fell 11% to £11,589 [7]
• Rightmove's AI-enabled Online Agent Valuation tool contributed to a c.50% increase in unique valuation leads sent to estate agents in H1 2026 (Rightmove plc) [1]

How dominant is Rightmove in UK property search in 2026?

More dominant than it was a year ago, on both measurement systems Rightmove reports. Its share of time spent across the major UK portals reached 90% on Comscore and 75% on SimilarWeb/Sensor Tower in June 2026 [1].

Those two numbers describe the same market and disagree by 15 percentage points. Comscore's panel counts Rightmove sites against a custom list of zoopla.co.uk, primelocation.com and onthemarket.com. SimilarWeb and Sensor Tower blend web and app estimates differently.

The direction of travel is consistent. In December 2024 the Comscore figure was 83%. By December 2025 it was 89%, and by June 2026 it was 90% [1][2].

Sources: Rightmove plc H1 2026 and FY 2025 results [1][2].

Note the awkward bit. Minutes fell 8% year on year in H1 2026 whilst share of time rose. Rightmove attributes this to a broad reduction in UK property browsing rather than anything specific to its own platform, and 8.4 billion minutes is still ahead of the 8.3 billion recorded in H1 2024 [1].

Portal dominance is not softening. If you are trying to reach a UK home mover, roughly nine out of ten minutes of portal attention sit with one company.

What does a portal actually cost an estate agency branch?

Rightmove's average revenue per agency advertiser was £1,636 a month in H1 2026, up £116 or 8% on the same period in 2025 [1]. Annualised, that is roughly £19,600 per branch per year.

Growth is coming mostly from products rather than price rises on the base package. Average products per agency branch rose 14% between H1 2025 and H1 2026, and 36% of independent agents now subscribe to the top "Optimiser Edge" package [1].

That distinction matters. An agent who feels squeezed by portal costs is often not being charged more for the same listing. They are being upsold featured slots, premium listings, valuation lead products and micro-sites, each individually justifiable and collectively expensive.

Set that against branch output. Propertymark members agreed an average of 8.0 sales per branch in May 2026 [5]. On GetAgent's average fee of £3,966 [9], Rightmove alone would represent a low single-digit percentage of a typical branch's sales revenue, before Zoopla, OnTheMarket, local press and its own website.

The comparison flatters larger agencies. A branch agreeing eight sales a month in Yorkshire and a branch agreeing eight in Kensington pay Rightmove similar money and earn wildly different fees. Foxtons recorded average revenue per branch of £2.654m in 2025 [7], for which £19,600 is a rounding error.

How many estate agency branches are there in the UK?

Rightmove carried 16,591 agency branches at 30 June 2026, up 206 or 1% since December [1]. Propertymark separately describes itself as representing over 19,000 members across over 12,500 branches [5].

Neither is a census. Rightmove counts paying advertisers, so it excludes agents who have left the portal. Propertymark counts voluntary members of a professional body.

What the Rightmove series does show is that the "high street agent is dying" narrative is not supported by portal membership data. Branch numbers rose 2% in 2025 and another 1% in the first half of 2026, helped by what Rightmove calls increased new agent formation [1][2].

The pain is elsewhere. New Homes developments on Rightmove fell 4% in the half and 6% year on year, with new developments coming to market at their lowest rate on record [1]. Developer marketing, not agency marketing, is where the contraction is happening.

How long does it take to sell a house in the UK in 2026?

Sixty-two days to secure a buyer nationally in June 2026, on Rightmove's measure, against 60 days in May and 81 days in January [3]. That is time from listing to a sale being agreed, not to completion.

The regional spread is enormous and it is the most useful table in the whole dataset for anyone planning local marketing.

Source: Rightmove House Price Index, July 2026 [3].

Scotland sells in 29 days. London takes 70. A national "average time to sell" figure in agency marketing is close to meaningless, and Scottish agents in particular should use their own number.

Then there is the second half of the transaction. Propertymark found 40.4% of member agents reported most of their agreed sales taking over 17 weeks from offer acceptance to exchange in May 2026 [5]. Add that to 62 days of marketing and the honest answer to "how long will this take" is most of a year.

How many viewings does it take to agree a sale?

Propertymark members recorded 2.2 viewings per available property in May 2026, with 44 properties on the books per branch and 8.0 sales agreed [5]. That arithmetic implies roughly 97 viewings across a branch's stock to produce eight agreed sales, or about twelve viewings per agreement.

Treat that as an order of magnitude rather than a precise ratio. The properties being viewed in a month are not the same properties being sold in that month, and the survey covers around 100 sales agents.

The pricing data explains a lot of the friction. In May 2026, 84% of Propertymark agents reported properties achieving less than the asking price, and only 11% reported properties achieving the asking price [5]. TwentyCi found the average newly listed UK property was priced 11.6% above its automated valuation model estimate in Q2 2026, up sharply from 5.7% a year earlier [6].

That is the single most commercially important number in this article. Sellers, or the agents pitching to them, are asking more above independent valuation than at any point in the recent series, in a market where sales agreed fell 5.8% year on year [6]. Overpricing is a marketing problem disguised as a valuation problem, and it is getting worse.

How do sellers choose an agent, and why is the valuation request the whole game?

The valuation request is the acquisition event. Everything an agency spends on brand, boards, leaflets and local SEO exists to produce one, and portals have worked out that this is where the money is.

Rightmove's AI-enabled Online Agent Valuation product, which it describes as its fastest-ever revenue growth from a new product, contributed to a c.50% increase in unique valuation leads sent to estate agents in H1 2026 [1]. Propertymark members conducted an average of 20 market appraisals per branch in May 2026 [5].

The shape of the funnel is clear. Twenty appraisals produce a branch's stock; that stock produces about eight agreed sales a month. Anything that lifts the top of that funnel is worth more than anything improving conversion further down.

It also explains why portals are moving into the space. A portal that owns the valuation request owns the introduction to the seller, the part of the relationship agents have historically guarded most closely.

Is lettings a better marketing bet than sales?

On demand density, yes, comfortably. Propertymark's May 2026 data shows lettings branches handling far more inbound enquiry against far less stock.

Source: Propertymark Housing Insight Report, May 2026 [5].

A lettings branch fields 98 registrations against twelve available properties. A sales branch fields 64 against 44. The lettings problem is stock acquisition, not demand generation, and marketing spend aimed at tenants is largely wasted.

The revenue case is stronger still. Foxtons Group derived 64% of its 2025 revenue from lettings and 30% from sales, with lettings contribution margin at 74.7% against 45.4% for sales [7]. Revenue per lettings transaction was £5,524 and rose 1%. Revenue per sales transaction was £11,589 and fell 11% [7].

Sales is the bigger single ticket and the more volatile one. Lettings is the annuity. Foxtons held 6.7% lettings market share in its core London markets in 2025, up from 6.2%, whilst sales share was broadly flat at 4.8% [7].

TwentyCi adds a supply-side wrinkle. Build-to-rent listings rose 22% in Q2 2026 alone, and June 2026 was the first year-on-year increase in available rental stock in five years [6]. Institutional landlords advertising direct to tenants are a genuine competitive threat to lettings agencies, not just a source of instructions.

Is AI search changing property portal traffic yet?

Barely, on the only hard number a UK portal has published. Rightmove reported that less than 0.5% of its traffic was referred from large language models in H1 2026 [1].

That figure deserves more attention than it has had. Rightmove is a Google Analytics-measured, 8.4-billion-minute platform with public disclosure obligations, and it is telling investors that LLM referrals are a rounding error.

The more interesting AI story is on-platform rather than off it. Rightmove's "Ask Rightmove" conversational search produced a c.40% uplift in average time on site and roughly double the propensity to send a lead among users who engaged with it, on data to 30 June 2026 [1]. The company had 46 strategic AI initiatives in flight at the half year, against 31 in December 2025 [1].

For agents, the implication is uncomfortable but simple. The AI layer that matters in UK property is being built inside the portal, on the portal's data, and it will be sold back to agents as a product.

What advertising rules do estate agents actually get caught by?

Comparative claims, mostly. A search of the ASA rulings database for "estate agent" in August 2026 returned seven formal rulings, all of them upheld, alongside fifteen cases resolved informally [12].

The most recent is instructive. On 10 June 2026 the ASA upheld a complaint against RE/MAX Property Specialists over a leaflet and Facebook post comparing its service with identifiable competitors, breaching CAP Code rules 3.1, 3.7, 3.32 and 3.34 [10]. Claims including "Personal social media campaign", "Personalised sale board for maximum marketing effect" and "Same day feedback on all viewings" were found to be misleading or unsubstantiated, and the comparison was not verifiable because consumers could not check it against competitors' actual packages.

The lead-generation layer has been caught too. In December 2023 the ASA upheld a ruling against GetAgent, finding that its "impartial recommendations" claim and its statement that agents "don't pay to appear" were misleading, because registered agents who paid fees received preferential positioning [11].

If you run comparison marketing against named local rivals, the rule to know is 3.34. You must be able to point a consumer at the source. A tick-box table on a leaflet is the format the ASA has repeatedly found wanting.

How to read these numbers

Propertymark's monthly report is based on responses from around 100 sales agents and 100 letting agents [5]. It is genuinely useful directional data and it is a small sample. Treat month-on-month wobbles as noise.

Rightmove's share-of-time figures depend entirely on which measurement system you accept. Comscore says 90%, SimilarWeb and Sensor Tower say 75%, and both are in the same investor announcement [1]. Neither is wrong; they measure different panels and different app treatments.

Asking prices and achieved prices are different animals. Rightmove's £372,359 is what sellers ask for newly listed homes [3]. Zoopla's £271,900 is an estimate of what homes are worth [4]. They are not comparable and should never be presented side by side as though they were.

Portal attribution claims need care. Rightmove's "7 out of 10 vendor instructions" comes from Street CRM data and its "8 out of 10 tenants" from an RLTS survey asking applicants how they found the property [2]. Self-reported attribution overstates the last thing a consumer remembers.

Finally, HMRC counts completions, not marketing activity. UK residential transactions totalled 1,213,800 in 2025 and 552,370 in the first half of 2026 [8]. That first-half figure is about 4% below H1 2025, but H1 2025 contained a March stamp duty deadline spike of 165,690 transactions that distorts any year-on-year comparison.

What this means going into 2027

Portal costs will keep rising through product, not price. Agency ARPA grew 8% whilst average products per branch grew 14% [1], which means the effective unit price is barely moving and the bundle is getting bigger. Budget for that shape.

Overpricing is the industry's biggest self-inflicted marketing wound. A 11.6% average gap between asking price and automated valuation [6], with 84% of agents reporting below-asking sales [5], is a pipeline of stale stock and price reductions waiting to happen.

Lettings deserves a bigger share of marketing budget than most sales-led agencies give it, and that budget should point at landlords rather than tenants. Eight applicants per available property [5] means tenant demand generation is close to worthless.

AI search is not yet a traffic problem for UK property, at under 0.5% of Rightmove referrals [1], but the agentic tools being built inside the portals will change what agents pay for by 2027. Rightmove has said agentic solutions for estate agency go live in H2 2026 [1].

And if you run comparative advertising, get your substantiation in order now. The ASA's 2026 ruling against RE/MAX is a template for how these complaints get upheld [10]. If you want help working out how findable your estate agency really is in search and AI answers, you can book a call with me.