Project 28, one year on: the first four weeks are the problem

What Project 28 is

Project 28 is a charter, not a law. Landmark Information Group brought together lenders, brokers, estate agents, conveyancers and data providers in September 2025 around one target: bring the time from sale agreed to exchange down to 28 days, from an average that was then over 100. Its eight commitments are practical rather than grand: instruct the seller's conveyancer at listing rather than after an offer, put key property information in front of buyers before they bid, use trusted digital data instead of paper, share documents securely, and make the chain visible to everyone in it. In March 2026 the Charter opened membership to organisations of any size.

One year on, the conversation has moved from commitment to delivery. The anniversary coverage describes upfront legal packs being piloted before marketing, property intelligence used to spot risk earlier, and a digital home-buying service designed to capture information once and reuse it across a transaction. The panel discussion turned on a single idea: bring the identity checks, the source-of-funds checks, the lender requirements and the risk indicators forward, so that issues surface before the sale is agreed and more of the work can happen in parallel.

Why the first four weeks matter more than the next twelve

The Charter's data partner is TwentyEA, and their numbers are the ones we use too. Nearly one agreed sale in four fell through in the first quarter of 2026, and 38% of those collapses happened within four weeks of the sale being agreed. Nearly 16% happened in the first two. At that point the conveyancing has barely started. What has actually happened is that a buyer nobody checked has been accepted, and the reasons they cannot proceed surface as soon as anyone looks.

That reframes the 28-day target. Speeding up the legal work is necessary, and the Charter members are doing it. But a fast process that starts with the wrong buyer is a fast fall-through. The first four weeks are not a conveyancing problem; they are a qualification problem, and the fix has to sit before acceptance, not after it.

What "bring the checks forward" looks like in practice

This is the part of the Charter HomeBid was built for. On HomeBid a seller cannot accept a buyer who has not passed. Every bidder is scored out of 100, with verified facts and declared ones labelled separately. Identity is checked to a standard aligned with HMLR Practice Guide 81. Source of funds and source of wealth are captured with the evidence, because a bank balance is not a source. PEP and sanctions screening runs automatically, blocks acceptance on a hit, and refreshes daily while the reservation is live. The buyer names their solicitor before acceptance, and it counts in the score.

Then the second half of the Charter's ask, trusted and shareable data, becomes concrete. At acceptance HomeBid generates a digital twin of the transaction, built to the Property Data Trust Framework, an open standard, and delivers it to both conveyancers within 48 hours. The property, the parties, the verified identity and the evidence behind it, the financing position, the professionals appointed, and the acceptance record, in JSON their systems can load and a PDF they can read. It is an accelerant for their due diligence, not a substitute for it.

Where we are honest about the gap

The Charter is written for resale chains. HomeBid launches with housebuilders, in late October, where the same checks apply at the point of reservation and there is no chain to manage. That is a deliberate choice about where to start, not a view about where the problem ends: the estate agent, the broker, the surveyor and the compliance providers are all in the transaction and all in the twin, and the resale journey follows the new-build one. Nor are we claiming outcomes yet. Our targets, every accepted reservation fully verified, the package with the panel firm within 48 hours, first-four-week cancellations halved against the developer's own baseline, are design targets until live transactions prove them. We will publish the measurements when we have them, against the baseline, not against a slide.

What we would add to the conversation

• Verification belongs at the point of offer, on every bidder, not at the point of instruction on the one who won. A seller choosing between bids should be choosing between verified positions.
• A declared fact and a verified fact should never look the same on screen. Most fall-throughs live in the gap between the two.
• The package should be an open standard. The value is in having the verified data at reservation; the format should be something any conveyancer, lender or settlement rail can consume without asking permission.
• Measure against the seller's own baseline. Industry averages make good headlines; the number that changes behaviour is a sales director's own cancellation rate, before and after.

Project 28 has done something the industry had not managed for a long time: put the people who each hold one piece of the transaction in the same room and agree what "better" means. Our contribution is a narrow one, verify the buyer first and hand everyone else a clean start, and we would rather be measured on it than talk about it.

Further reading

• Project 28 Charter: the eight commitments
• TwentyCi: why the Charter was formed
• The Intermediary: the first anniversary
• MoneySuperMarket: what Project 28 means for buyers and sellers
• How HomeBid works, in five steps

Statistics: TwentyEA Property and Homemover data, reported April and August 2026. Descriptions of Charter activity are drawn from the public coverage linked above; HomeBid is not a Charter member and does not speak for it.