Upfront information will not fix most fall-throughs

Two sets of figures about the second quarter of 2026 came out separately, and they are considerably more interesting read together than apart.

The first counts the damage. 71,959 property transactions collapsed in that quarter, 6.6 per cent more than in the first, at an average wasted cost of £3,584 a time. That puts the total at £257.9 million across three months. Those are TwentyCi figures analysed by House Buyer Bureau.

The second counts the causes. Quick Move Now's breakdown of why sales failed in the same quarter: a buyer refused a mortgage or lending falling through, 33 per cent. Survey issues, 27 per cent. A break in the chain, 13 per cent. A change in the buyer's circumstances, 13 per cent. Legal issues, 13 per cent.

Hold those two lists side by side, because the second one ought to govern how anybody talks about the first.

What is being promised

The government's home buying and selling reform roadmap makes its case on exactly this problem. It puts the cost of fall-throughs at "circa £400 million per annum" to consumers, cites independent research putting the cost to the economy at around £1.5 billion a year, and notes that a transaction "takes around 120 days on average to complete, and the journey is now around 60% longer than it was in 2007".

Its answer, eventually, is sales packs prepared before listing, digital logbooks, and a defined minimum of standardised data. We wrote about what that means for the information an agency holds in the home buying reforms as a data project, and we think that shift is real and worth preparing for.

What we are much less sure about is the claim doing the rounds underneath it: that upfront information is the fix for fall-throughs. The causes data does not support it, and the people repeating it loudest tend to be selling something.

The uncomfortable arithmetic

Sixty per cent of failed sales in that quarter came down to lending and survey. Neither of those is a disclosure problem.

A sales pack does not tell a lender whether to lend. It has nothing to say about a buyer's income, their credit file, the rate they were offered in March and can no longer get in June, or a down valuation on a property whose asking price the market has moved away from. That is a third of all failures, and the most complete pack in the world leaves it exactly where it was.

A survey happens on the actual building. It finds what it finds. Some of what it finds was knowable and could have been disclosed at the start, and we will come back to that, but a great deal of it was not knowable to anybody until a surveyor put a ladder against the wall.

So what is left for upfront information to work on? The legal 13 per cent, most obviously, and some unknown share of the survey 27 per cent where a defect somebody already knew about turns up at week eight instead of week one.

Take the legal slice on its own, because it is the one that is cleanly in scope. Thirteen per cent of 71,959 is roughly 9,400 sales. At the average wasted cost, that is somewhere around £34 million in a quarter.

That is real money, and it is worth going after. It is also about an eighth of £257.9 million. Anybody telling you that upfront information solves the fall-through problem is describing a fraction of it as though it were the whole.

Two caveats on that sum, because we would rather state them than have somebody find them. The volume and cost figures and the causes breakdown come from two different studies of two different samples, so combining them gives an indication rather than a measurement. And applying the average wasted cost to the legal category assumes a legally-doomed sale costs about what any other failed sale costs, which nobody has demonstrated either way.

Why the big two are out of reach

It is worth being clear about why, because it is not a failure of the policy.

Lending failures are a function of the buyer and of the market between offer and completion. The reform has a separate answer to those, which is to shorten the window: a transaction that takes eight weeks rather than seventeen gives a buyer's circumstances less time to change and a rate less time to expire. That is a genuine benefit and it is a speed argument, not a disclosure one.

Survey failures are a function of the building. Upfront information helps at the margin, where the seller or agent already knew about the thing the survey later found. It does nothing about the rest.

Which means the honest version of the reform's case is not "this stops sales collapsing". It is closer to "this removes one category of collapse, shortens the window in which the others can happen, and makes the whole thing less miserable". That is a decent case. It does not need the oversell.

What a listing can actually do

Here is the part that is genuinely an agent's to control, and it is smaller than the marketing suggests and larger than nothing.

Everything in the legal 13 per cent is, by definition, a fact about the property that existed before it was listed. A restrictive covenant. A shared access with no maintenance agreement. An unregistered extension. A lease with an unhelpful number of years left, a service charge that has doubled, a ground rent clause a lender will not accept. None of those appear between offer and exchange. They were all true on day one, and in most cases somebody knew.

The question worth asking about your own stock is not whether you could produce a sales pack. It is narrower and more awkward: on the properties that fell through this year on a legal point, did anybody at the agency know about that point when the instruction was taken?

If the answer is sometimes yes, that is the bit within reach, and it costs nothing to change. It is a question set at valuation and a place on the record to put the answers, which is the same underlying problem as material information and the same reason we keep going on about property facts being held as fields rather than as sentences.

If the answer is genuinely no, nobody knew, then upfront information would not have saved those sales either, and you can stop worrying about that particular sales pitch.

Where we land on it

Fall-throughs are getting worse, the quarterly cost is rising, and the reform aimed at them is years from being law. In the meantime a fair amount of product is going to be sold on the back of those numbers.

Our position is that the direction of travel on standardised, machine-readable property information is right and worth building for, which is what we argued in the piece linked above. It is simply not the fall-through cure, and pretending otherwise sets an expectation that the data will not meet. When agents buy a thing on a promise it cannot keep, what they conclude a year later is that the whole idea was nonsense, and that is worse for everybody than an accurate pitch would have been.

Go after the thirteen per cent. It is worth having. Do not pay for it as though it were the whole £257.9 million.

A note on sources. The transaction volume, the 6.6 per cent quarterly rise, the £3,584 average and the £257.9 million total are House Buyer Bureau's analysis of TwentyCi data for the second quarter of 2026. The causes breakdown is Quick Move Now's, for the same quarter. The roadmap figures and quotations are from the government's home buying and selling reform roadmap. Where we have combined the two datasets we have said so and said why it is indicative.