London-wide leak specialists
ResidentialCommercial
Get in touch
Book an engineer

What 25 London Leak Investigations Actually Found

Across 25 leak investigations in our own job records between 17 July and 18 September 2026, a failed pressurised pipe was the cause nine times. Six were a fitting, valve or sanitaryware item reachable without opening anything, five found no plumbing leak at all, and the remaining five were drainage, tile grout, a boiler component, or unresolved on the day.

Leak Fix LondonLast updated

A Leak Fix London engineer working on pipework in a London home

The counts

Twenty-five investigations, each assigned one primary cause from the written finding:

What it turned out to beInvestigations
A failed pressurised supply or heating pipe9
A failed fitting, valve or sanitaryware item6
No plumbing leak found at all5
A waste or drainage defect, not a pressurised pipe2
A boiler component, with the pipework sound1
Tile grout, not plumbing1
Not established on that visit1
Total25

The number people expect to be large is the first row, and it is the minority outcome. In sixteen of twenty-five, the thing that was wrong was not a pipe.

Method, so you can discount it properly

The source is our own job management system, read on 19 September 2026. It holds 149 job records, 85 of which carry a date, running from 4 April to 18 September 2026.

From that, we took every job booked as a leak, damp, water ingress or unexplained water loss investigation where the record states what was found. That is 25 investigations. Twenty-one carry a full written report with a findings and cause section; four carry a shorter findings note. All 25 fall between 17 July and 18 September 2026, a nine-week window, because records before mid-July are one-line ledger imports that name the job but not the outcome.

Excluded: jobs with no recorded outcome; two leak bookings cancelled before attendance; and all the plumbing, heating, drainage, electrical, appliance and building work in the system that was not booked as a leak investigation. Where one property took more than one visit, the visits are counted as one investigation. Each investigation is assigned a single primary cause, taken from the engineer's stated conclusion rather than from the booking.

The limits are worth being blunt about. Twenty-five is a small number. Nine weeks of London summer contains no winter, so nothing here says anything about frozen or burst pipes. The properties are self-selected by who telephoned us, and the mix leans towards managing agents, landlords, tenanted flats and blocks rather than owner-occupied houses. This is one firm's caseload, not a survey.

The nine pipe failures

These are the jobs the word "leak" usually conjures, and they were the minority:

  • A pinhole in a 15mm plastic pipe inside a ceiling void, discharging under mains pressure and coming through a bathroom ceiling around a light fitting.
  • A 15mm cold supply to a toilet, failed behind the pan connector where an access board hid it.
  • A 22mm heating pipe under a floor, perforated by rodent gnawing, in a cupboard beneath the stairs.
  • A hole in a lead supply pipe under a hallway floor, repaired with a clamp because the pipework stacked above it made a cut-and-replace substantially more invasive.
  • Pinholes in both the hot and the cold concealed pipework of the same bathroom, between the bath and the toilet.
  • A joint between old imperial-sized plastic underground pipe and a stopcock, below the floor of a large hall.
  • A buried incoming supply pipe between the external meter and the internal stopcock, roughly a metre down. Confirmed by elimination; the exact point was not located on that visit.
  • A concealed hot-water circuit. The cold held pressure for 30 minutes and the hot dropped immediately. The circuit was identified; the point was not, and tracer gas was the recommended next step.
  • An elbow on a toilet supply under a hallway floor, found after the boards above it warped.

Two things stand out. Only two of the nine were in a wall or ceiling; six were under a floor or in the ground. And in two of the nine, the leak was proved to exist but not pinpointed on the visit that proved it.

The six fitting failures

These cost far less to fix than the pipe failures, and several had been running for a long time:

  • A loft cold-water storage tank ball valve and a WC filling valve, both passing water continuously. Found on a job booked because the water bill had roughly doubled.
  • Two WC cisterns passing, one badly, plus a dripping bath tap. Found on a job booked from a water company notice of about 100 litres a day.
  • A shower waste connection that had lost its seal, with perished silicone round the tray as a second route for the same water.
  • A bath tap whose diverter had failed internally.
  • A bath waste pipe that had come away from its supporting brackets, so the joint opened every time the bath drained.
  • Water spilled during another trade's replacement of a concealed cistern, combined with a ball valve seized solid with limescale.

Four of those six are a toilet, a tank valve or a tap. If you are reading this because your bill has jumped, start there. Reading your meter with everything turned off costs nothing and is the single most useful thing you can do before anyone attends.

The five where there was no leak

Five of the twenty-five found no leak in the plumbing at all. A sixth found no leak in the pipework but a failed boiler component.

  • Mould in several rooms of a flat. Hot and cold both held pressure for fifteen minutes; the waste system was run and watched; every accessible cavity was dry. Attributed to historic ingress or condensation.
  • Damp reported in a property for years, with previous engineers unable to find a source. Hot, cold and heating were pressure tested and then tracer gas tested in turn. All three held.
  • A water company notice of extra consumption, investigated over three visits with visual inspection, pressure testing, thermal imaging, an inspection camera and tracer gas. Nothing. The supply turned out to be shared with the flat above, and the discharge was theirs.
  • Damp marks inspected with an endoscopic camera, with no leak visible. Either a past leak already resolved, or ingress through an external wall.
  • A constant dripping sound behind a wall, reported as a possible leak inside a concealed cistern. The cistern tested sound; the noise was condensation running down the inside of it.

Four of those five started as damp, mould, staining or a sound. None of them started as visible running water. That distinction is the most useful single thing in this dataset, and it has its own write-up.

The four that were something else entirely

Two were drainage rather than supply: a hole in an external four-inch soil pipe at a 45-degree junction, which only wet the brickwork when the bathroom above was used; and a blocked gully outside, which was soaking masonry and putting water into a cellar. That second one was booked as tracer gas detection, and the blockage was cleared instead.

One was tile grout. The bottom row of tiles in a shower had lost its grouting, so water ran behind them and onto the ceiling below every time the shower was used. It was confirmed by asking the occupier to run the shower and watching.

One was a boiler component. The system would not hold pressure, tracer gas was run twice and found nothing, and on re-pressurising, water came out through the pressure relief valve. That points at the expansion vessel rather than the pipework, and it is a different job with a different trade at the end of it.

What this changes about how you should book

If the pattern here holds, the most expensive mistake is booking the fix before anyone has established the fault. Six of the twenty-five needed a washer, a valve or a cartridge. Five needed nothing. Eight needed a floor, wall, ceiling or the ground opened. Those are three completely different jobs, and you cannot tell which one you have from the symptom.

It also explains why detection is charged by the hour rather than as a fixed fee for a find. Our rate is £150 per hour, agreed before we attend, and under our no find, no fee terms you do not pay for the time spent looking if we do not find it. On the five where nothing was found, that is exactly what happened.

When not to call us

If your only symptom is mould in a corner, on a cold external wall, or around a window, and no fitting is dripping and the meter is still with everything off, a damp surveyor is a better first call than a leak engineer. Two of the five no-leak outcomes above are jobs that a damp specialist should have had first.

If your boiler loses pressure and the discharge pipe outside drips after it fires, read the pressure article before booking detection. That symptom points at the vessel or the relief valve, and neither is found with tracer gas.

And if a tap, a cistern or a visible joint is dripping, that is a plumber's hour, not a detection job. Nobody needs to survey a property to find a tap you can already see.

Frequently asked

Is 25 jobs enough to conclude anything?
It is enough to describe what we found and not much more. Twenty-five is a small sample from one firm over nine weeks, self-selected by who called us. Treat the direction of the findings as informative and the exact proportions as rough. We have published the method and the exclusions so you can judge it yourself.
Why does the period start in July if the records go back to April?
Records before 17 July 2026 are one-line ledger imports that name the job but not the outcome. A job titled "leak detection" with no findings cannot be classified, so including it would have inflated the sample without adding information.
Does this mean most leaks are not serious?
No. It means most reported symptoms do not turn out to be a failed pipe. The nine that were pipe failures included one discharging under mains pressure into a ceiling void and one losing around 420 litres an hour under a floor, both of which were doing real damage.
How were the categories decided?
By hand, from the engineer’s stated cause in the written report, not from how the job was booked. Each investigation was given one primary cause. Where a job had a secondary defect as well, such as perished silicone alongside a failed waste joint, only the primary cause was counted.
Do you publish any customer details from these jobs?
No. No name, address, postcode, job reference or property identifier appears in any of these posts, and individual jobs are described only in terms that would fit hundreds of London properties.

No find, no fee · £150 per hour. The rate is fixed before we attend and does not rise for a difficult property, an awkward access or an evening visit. The total depends on how long the job takes; the rate cannot change.