49% of Homes Withdrew Unsold in July: What Sellers Do Now

By Lisa Hayes · 24 August 2026

In July, 79,600 UK homes exchanged contracts and 76,300 came off the market unsold. That means only 51% of the properties that left estate agents' books actually sold — the other 49% were withdrawn, according to market data published by Property Industry Eye. The long-run average is closer to 58%, so roughly one in every fourteen sellers who would normally have got there this summer didn't.

If you're planning to sell this autumn, that figure tells you more about your odds than any house price index will. Prices are the story the papers run. Whether your house sells at all is the story that affects your life.

• 51%of homes leaving agents' books in July exchanged
• 76,300homes withdrawn unsold in July 2026
• 767khomes on the market on 1 August
• 9.9%gap between asking prices listed and asking prices that sell

What actually happened in July?

The data comes from Property Industry Eye's weekly market analysis by Chris Watkin, published on 21 August 2026 and covering week 32 (the week ending 16 August), along with the July completion figures. Two numbers sit at the heart of it.

The first is exchanges: 79,600 homes in July got all the way through to exchange and completion. The second is withdrawals: 76,300 homes were pulled off the market without selling. Watkin notes both figures will drift upward through August as more transactions are reported into the system, but the ratio between them is what matters, and that ratio is roughly one to one.

Put plainly: last month, a home leaving an estate agent's books was almost as likely to be a failure as a sale.

The seven-year average for that same measure is 57.6%. That average includes the frantic post-lockdown period when almost anything with a roof sold, so it's flattering. Even allowing for that, July's 51% is a meaningful step down. It's the quiet statistic behind every conversation you've had with a neighbour who "had it on for months and took it off in the end".

What does "withdrawn unsold" actually mean?

It's worth being precise, because the word covers several very different situations.

A withdrawal is recorded when a property is taken off the market by the seller or the agent without having exchanged contracts. Some of those sellers have decided not to move after all. Some have had a change of circumstance. Some are switching agents and will reappear next month under a different board. And a large number have simply run out of patience after months of viewings that went nowhere.

What a withdrawal is not is a repossession or a distressed sale. There is no forced selling in these figures. It is, overwhelmingly, the sound of people quietly giving up, and that is a very different signal from a market in trouble. It tells you the market is illiquid at current asking prices rather than collapsing in value.

One more caveat worth stating plainly: both the exchange and withdrawal figures for July will rise over the coming weeks as more transactions are reported through the system. The ratio is unlikely to shift much, but the absolute numbers will. If you see a slightly different figure quoted in a month's time, that's why.

Why are so many homes coming off unsold?

Not because buyers have disappeared. That's the assumption most sellers make, and it's the wrong one.

There were 24,400 homes sold subject to contract in week 32, up from 21,800 the week before and only slightly below the ten-year average for that week of 25,200. Year to date, 786,000 UK homes have gone sold STC. That is 7.1% down on 2025's 846,000, but it is 10.8% above 2023 and 6% above the pre-Covid 2017–19 average of 742,000. The decade average for this point in the year is 784,000. We are, almost exactly, on the ten-year norm.

So buyers are transacting at broadly normal volumes. The problem is that far more sellers are competing for them. There were 767,000 homes on the market on 1 August, against 763,000 twelve months earlier and 760,000 the month before. New listings for 2026 stand at 1.166 million year to date, 10.5% higher than the 2017–19 average of 1.059 million. Rightmove separately reported that the supply of available homes in July was close to a twelve-year high for the time of year.

Normal demand plus swollen supply gives you exactly what we're seeing. Enough sales to keep the market functioning. Not enough to clear the shelves.

Is this as bad as it sounds?

It's serious, but it is not a crash, and anyone telling you otherwise is selling something. Here's how the current numbers sit against their benchmarks.

Measure	Latest	Benchmark	Read
Homes leaving agents' books that exchanged	51% (July 2026)	57.6% seven-year average	Materially worse
Homes sold STC, year to date	786,000	784,000 decade average	Bang on normal
Net sales, year to date	609,000	585,000 (2017–19 average)	4.2% ahead
Sell-through rate	14.2%	15.5% pre-Covid	Slightly soft
Fall-through rate	25.7%	24.5% decade average	Marginally worse
Homes for sale reduced in price	13.7% (July)	11.2% six-year average	Elevated
Average £ per sq ft, agreed sales	£345.41 (July)	£341.43 a year ago	Up 1.2%

Read that table properly and a clear picture emerges. Transaction volumes are normal. Achieved prices are slightly up, not down: £345.41 per square foot on agreed sales in July, 1.2% higher than a year ago and 11.9% higher than five years ago. Watkin argues this per-square-foot measure tracks the eventual Land Registry index with about 98% accuracy, roughly five months ahead of publication. If he's right, actual sold prices are still creeping up.

What has broken down is the matching process. Too many homes are listed at prices that the available buyers won't pay, so they sit, then they reduce, then they withdraw. The market isn't refusing to buy. It's refusing to overpay.

The gap that explains almost everything

This is the single most useful figure in the whole dataset, and almost nobody reports it.

The average asking price of homes currently listed for sale is around £392,000. The average asking price of homes that actually go sold subject to contract is around £357,000. That's a gap of 9.9%.

Historically that gap has run at 16% to 17%. It has narrowed sharply, and there are two ways to read it. The generous reading is that sellers have wised up and are pricing more realistically from day one, which is what Rightmove has also observed. The blunter reading is that the pool of properly priced homes is now so much smaller than the pool of listed homes that the two averages have converged out of sheer attrition — the overpriced stock has stopped moving at all, so it no longer influences the "sold" figure.

Both readings lead to the same conclusion. There is now a fairly narrow band of price at which a home sells in 2026, and everything above it stalls.

Here's the stat that proves it, and it surprised me when I read it: four out of five homes listed and sold so far in 2026 achieved that sale without a single price reduction. Eighty per cent of successful sales were priced right at launch. Meanwhile 13.7% of all homes on the market were reduced during July alone, and 22,500 were cut in one week of August.

Reductions are not the route to a sale. They are the symptom of a launch price that was wrong, and by the time you're reducing, you've usually already burned the weeks that mattered most.

Why the first three weeks decide your sale

Every serious buyer searching your area has portal alerts. When your house goes live, it lands in front of the entire active buyer pool at once. That is the moment you have maximum attention and maximum bargaining power.

If the price is 8% too high, those buyers filter you out and never look again. You don't get a second first impression when you reduce three months later. Worse, portals display how long a property has been listed, and buyers use that as a haggling tool. A home that's been up for four months invites a low offer in a way an identical home listed last Tuesday does not. If you want to see how that plays out, our guide on how long a property has been on the market walks through what buyers actually infer from listing age.

The withdrawal data is the aggregate consequence of thousands of sellers learning this the expensive way.

What about mortgage rates and the Budget?

They matter, but less than the pricing story, and not in the direction most people assume.

The Bank of England held Bank Rate at 3.75% on 30 July 2026, with six MPC members voting to hold and three voting for an increase to 4%. Rate has now been unchanged since December 2025. What has moved is the mortgage market itself, which prices off swap rates rather than Bank Rate directly. Moneyfacts data put the average two-year fixed deal at around 5.55% in mid-August, and lenders have been repricing in both directions: HSBC and Barclays have raised rates in recent weeks while Halifax and Nationwide have trimmed theirs by up to 0.15 percentage points.

That volatility is corrosive in a way that a simple high rate isn't. Buyers who can't predict what a deal will cost in six weeks hesitate. Hesitation lengthens chains, and longer chains mean more fall-throughs, which is part of why the fall-through rate sits at 25.7% against a decade average of 24.5%.

Then there's the new Chancellor's first Budget in the autumn. Speculation about property taxation has been running hot for months. Rightmove cited exactly this cocktail — geopolitical uncertainty, a shifting mortgage rate landscape and the Budget — when it downgraded its 2026 house price forecast from +2% growth to a range of 0% to -2%. Zoopla has trimmed its own forecast to a 1% rise.

Read the forecasts for what they are. Both portals are saying the same thing in different registers: nominal prices will end 2026 roughly where they started, and the action is all in volumes and timings rather than headline values.

What does this mean if you're selling?

If you're a realistic seller with a decent property in a normal price bracket, this is a workable market. If you're hoping to test a hopeful number "just to see", the odds have moved against you sharply, and the cost of finding out is four months of your life.

What should you do now?

Get more than one valuation, and interrogate the highest one. Some agents win instructions by flattering you. The valuation that gets you on the books is not the same thing as the price that gets you sold. Ask each agent for evidence: three comparable properties that have exchanged in the last three months, not three that are still sitting there asking. Our how much is my house worth guide covers what a defensible valuation should contain, and free house valuation explains where the online estimates go wrong.

Price to the sold-STC average, not the listed average. That 9.9% gap is your map. Look at what's actually going under offer in your postcode and price into that band, not above it.

Assume you launch once. Photography, floorplan, description and price all need to be right on day one, because the portal alert only fires once.

Get your paperwork ready before you list. With the fall-through rate at 25.7% and searches taking as long as they currently do, anything you can prepare in advance protects your sale. Title deeds, guarantees, building regs sign-offs, FENSA certificates, lease documents if you're leasehold. Our conveyancing guide sets out what a solicitor will ask for.

Vet your buyer properly. In a market where a quarter of agreed sales collapse, the highest offer from an unverified buyer in a four-link chain is worth less than a slightly lower offer from someone with a mortgage in principle and nothing to sell. Ask the agent to prove funds and position before you take the house off the market.

Decide honestly how much certainty is worth to you. If your move depends on a fixed date, a job relocation, a divorce settlement or a probate deadline, the open market's 51% success rate may not be a risk you can carry. Selling to a cash buyer means accepting below market value, and you should go in with your eyes open about that: our guide to what percentage of market value cash buyers pay gives you the honest numbers, and cash house buyers explains how the process runs. It is the right answer for some people and a bad answer for others. The mistake is choosing without comparing.

What if your house has already been sitting there?

You are, statistically, in very ordinary company. Roughly 76,000 households a month are in the same position.

The instinct is to blame the market. Sometimes that's fair. More often the honest diagnosis is one of three things: the price is wrong, the presentation is wrong, or the property has a specific issue the marketing hasn't addressed.

Work through it in that order. If you've had plenty of viewings and no offers, the price is close but the property isn't converting, which usually means condition, layout or something visible on the day. If you've had very few viewings at all, the price is wrong and the photographs probably are too. If you've had offers that keep collapsing, the problem is buyer quality or something surfacing in survey.

Withdrawing and relisting in the spring is a legitimate strategy, but only if you change something. A property that comes back at the same price with the same photographs is recognised instantly by the same buyers who passed on it. Our guide on how to sell a house that won't sell goes through the diagnosis properly, and if the issue is that the property needs work, selling a house that needs updating covers your options.

Should you let it out instead?

It's the question I get most from sellers who have been on the market since spring, and the rental numbers are genuinely inviting at first glance. Average UK rent stood at £1,805 pcm in August 2026, and 135,928 new rental properties came to market in July, against 128,821 in July 2025. Available rental stock was 323,000 in July, up from 319,000 a year earlier.

Look closer, though, and the rent figures are flat, not booming. August 2026's £1,805 compares with £1,800 in August 2025, a rise of well under 1% across the year, and the year-to-date average sits at £1,765. Rents are no longer doing the heavy lifting they did between 2021 and 2024, when the same measure was £1,394.

Letting also carries costs most accidental landlords underestimate: consent to let from your lender or a switch to a buy-to-let product at a higher rate, compliance obligations, agency management fees, void periods, and the capital gains position when you eventually do sell a property that's no longer your only or main residence. Our guide to capital gains tax when selling your home explains why the timing of that decision matters more than people expect.

Letting to avoid a price reduction is usually a way of paying more, over a longer period, to arrive at the same answer. Letting because you actually want to be a landlord is a different decision entirely, and a legitimate one.

Where does this data come from?

Fair question, and one you should ask of any property statistic. These figures come from Property Industry Eye's weekly market analysis, compiled by Chris Watkin from listing, sale-agreed, exchange and withdrawal records across the UK market, published on 21 August 2026 and covering week 32 plus the July completion data.

That makes it different in kind from the indices you see in the headlines. Rightmove measures asking prices at the point of listing. Nationwide and Lloyds measure mortgage-approved purchase prices from their own lending books. The ONS and Land Registry measure completed sales, accurately but with a lag of several months. None of them count the homes that failed to sell, which is precisely why the withdrawal rate rarely makes the news despite being the number that decides whether your move happens.

What happens between now and Christmas?

Three things are worth watching, and I'd weight them in this order.

The Budget. This is the big one. Property taxation has been openly discussed all year, and a meaningful chunk of buyers and sellers are simply waiting to see what lands before committing. Expect the market to feel sticky until the Chancellor sits down, and expect a burst of activity afterwards regardless of what's announced, because certainty itself unlocks decisions. If you can transact before it, you avoid the queue.

Mortgage pricing. Oxford Economics expects Bank Rate to stay at 3.75% well into 2027. If that holds and swap markets settle, fixed rates should stabilise, and stable is more valuable to buyers than slightly cheaper. Watch the two-year and five-year averages rather than the headline best buys, which usually carry fees and loan-to-value requirements most people can't meet.

Stock levels. With 767,000 homes on the market and 487,000 in agents' pipelines (down from 508,000 a year ago), the supply overhang is the thing that will keep pressure on pricing. Autumn usually brings a fresh wave of listings in September. If you're going to market, going early in that wave beats going late.

My honest view: this is not a market that punishes sellers. It's a market that punishes optimistic pricing. Those are different things, and the 49% withdrawal rate is what happens when people confuse them.

If you're weighing up whether to go to market this autumn or take a guaranteed offer instead, the sensible first step is to see both numbers side by side rather than guessing. You can compare what your home would fetch on the open market against what cash buyers would pay, with no obligation either way. Knowing the gap is what lets you choose properly, and in a market where half of sellers are coming away with nothing, choosing properly is most of the battle.