Every building will, at some point, leak. A joint will fail, a flexible hose will perish, a washing machine connection will let go while nobody is home. This is not a design failure. It is the ordinary behaviour of pressurised water over a long enough period of time.
What is a design matter is what happens in the hour afterwards — whether the water can be stopped in thirty seconds or forty minutes, whether the failed joint can be reached without a demolition saw, whether the damage stays inside a utility cupboard or travels through three floors of finishes. By the time the leak starts, almost all of that has already been settled, in a drawing produced months or years earlier by somebody who was probably thinking about something else.
Two Buildings, One Burst Pipe
Picture the same failure in two apartments: a split in a concealed supply line feeding a kitchen sink.
In the first, the occupier knows where the stopvalve is, because it sits behind a labelled hinged panel inside the kitchen cupboard. The water is off before the floor is covered. The pipework runs in an accessible service void, so the repair is a cut, a joint and a panel refitted. The affected area is a metre of cupboard base.
In the second, the supply enters somewhere beneath a bank of fitted units with no access panel. Nobody is certain where the isolation valve is, and the only stopvalve anyone can locate is in the street. By the time the supply is off, water has found the junction between a timber floor and a plasterboard partition, and the repair involves lifting flooring, cutting out joinery and a drying programme.
The same component, the same mode of failure, and two outcomes separated by a factor of perhaps fifty in cost and by months in disruption. Nothing about the plumbing differed. The drawings did.
The aggregate of those second outcomes is considerable. In the UK, the Association of British Insurers estimates that insurers pay around £1.8 million every day for domestic water damage claims, and Aviva’s own claims data puts the average escape-of-water claim at £11,273 across 2023 to 2025. Those are UK figures and other markets will differ, but the mechanism producing them does not.
What the Regulations Already Require
It is worth saying plainly that in some jurisdictions this is not merely good practice. In England and Wales, the Water Supply (Water Fittings) Regulations 1999 address it directly, and rather more firmly than many designers realise.
Schedule 2, paragraph 7 prohibits concealing a water fitting — specifically, ‘no fitting which is designed to be operated or maintained, whether manually or electronically’ may be a concealed water fitting. The same paragraph defines a concealed fitting as including one ‘in any other position which is inaccessible or renders access difficult’. Read together, those provisions say something quite pointed: a valve you have to dismantle joinery to reach is not a compliant detail. Paragraph 7 also prohibits embedding a water fitting ‘in any wall or solid floor’.
Paragraph 10 requires a stopvalve ‘conveniently located to enable the supply to those premises to be shut off’ and, where a supply is shared, ‘a stopvalve to which each occupier of those premises has access’. Paragraph 11 requires ‘an adequate number of servicing valves and drain taps’, and that ‘a sufficient number of stopvalves shall be installed for isolating parts of the pipework’.
Two honest caveats belong here. The regulations do not, in those words, require that every valve be readily accessible, and they contain no general requirement that pipework be accessible for inspection or repair. And this is one country’s framework; comparable regimes exist elsewhere under other names, and anyone working outside England and Wales should check their own. But the direction is unambiguous. Isolation is expected to be possible, in sensible places, by the people who will need it.
Why Concealment Became the Default
None of this is news to the people specifying it, which raises the obvious question of why so many buildings are so difficult to isolate.
The answer is largely aesthetic, and it is not unreasonable. A generation of residential interiors has been built on flush surfaces, uninterrupted planes and the absence of visible mechanism. An access panel is a line on a wall. A labelled valve is a piece of equipment in a room designed to contain none. Boxing services in is also, very often, the cheapest way to resolve a coordination clash discovered late.
Then the pressures compound. Services are frequently the last thing resolved, by which point the ceiling void has been given over to ductwork and the riser has been sized for everything except maintenance access. Concealment becomes the path of least resistance rather than a decision anybody actively took.
The result is a building that looks exactly as intended on handover day and grows steadily more expensive to own from that point onwards.
The Decisions That Govern a Failure
The useful move is to treat failure as a design case — much as means of escape is treated as a design case — and to ask a short set of questions while they can still be answered cheaply.
Where is the incoming stopvalve, and could somebody who has lived here for six months find it in the dark? Can each wet area be isolated independently, so that a fault in one bathroom does not mean shutting down the building? At every concealed junction, manifold and valve, is there a planned means of access — and is that access drawn, rather than left to a contractor’s judgement on site? Does the system have low points that can actually be drained?
And where do the wet areas sit in relation to the things most expensive to replace? Is the shower above the stairwell, the plant room over the archive, the kitchen directly above a double-height space with a plaster ceiling? That last question is a planning decision rather than a services one. Stacking wet rooms is routinely discussed as a matter of cost and gradient; it is also a matter of risk containment, and it costs nothing to weigh at concept stage.
The First Fifteen Minutes
There is a practical test for all of this, which is to imagine the person who turns up to deal with the failure.
They arrive at a building they have never seen, usually at an unhelpful hour, with no drawings and an occupier who is understandably not at their most precise. What they can achieve in the first fifteen minutes is determined almost entirely by what the design left them: a findable stopvalve, a labelled isolation point, a panel that comes away with a screwdriver rather than a chisel. In the UK, the emergency plumbers in London and their equivalents elsewhere work around the clock precisely because water keeps no office hours.
It is worth being realistic about response times. Attendance depends on location and severity, and no honest provider guarantees an arrival time. Which is rather the point: the minutes before anybody arrives belong to the building, and therefore to whoever designed it.
Retrofit and Period Buildings: Harder, and More Important
Everything above is easier in new build, where the question can be asked before anything is buried. In retrofit and conversion work it is harder, and it matters more, because the existing condition is usually undocumented and frequently worse than assumed.
In an older building there may be no record of where services run, valves painted shut for decades, and a water main whose route nobody has established. Before any of it can be designed around, it has to be found — and in a building with finished interiors or protected fabric, finding it without destructive opening-up is the whole problem. Techniques such as moling allow a mains supply to be replaced along a route beneath a garden or driveway without excavating its full length, which matters a great deal when the surface above is a finished landscape or a listed forecourt.
The design lesson is the same as in new build, only arriving earlier: establish what is actually there, decide where access will be, and only then resolve the finishes that will sit over it.
Handing Over a Building That Can Be Maintained
The final opportunity to influence any of this comes at handover, and it is the one most often wasted.
A set of as-built drawings recording where services actually run — rather than where they were intended to run — is worth more to a building across its life than almost any other document the project produces. Valves that are physically labelled, an isolation strategy explained in plain language rather than buried in an operation and maintenance manual nobody opens, and a simple plan showing where to turn the water off: these cost very little and shape how the building behaves for the next fifty years.
It is also the part of the service that separates a project which was finished from one that was merely completed.
Designing In the Assumption of Failure
Designing for failure is not pessimism, and it is not an argument against concealed services or clean interiors. It is an argument for deciding deliberately and early where access will be, and then designing the finishes around that decision rather than discovering it afterwards.
The components will fail; that much is fixed. What remains genuinely open — and genuinely a design question — is whether the failure costs a morning or a floor. Contractors who spend their working lives on the consequences tend to have the clearest view of the causes. Maintracts Services, a Gas Safe registered firm and a member of the APHC and CIPHE that has worked across London since 1974, covers plumbing, heating, drainage and electrical work on precisely the buildings described here. The recurring observation from that side of the industry is rarely that the plumbing was wrong. It is that nobody could get to it.
Draw the access. Label the valves. Then conceal everything else as beautifully as you like.
