Escape of Water or Water Ingress, and Why Insurers Care

Two ceilings can look identical. Same brown ring, same sag, same smell. One is a paid claim and the other is a maintenance bill. The difference is where the water came from, and your insurer decides that before it decides anything else.

Escape of water means water that has left a system built to contain it: pipes, tanks, appliances, a heating circuit. Water ingress means water that has come in from outside the building envelope: a roof, a gutter, a parapet, a window seal, a basement wall. It sounds like jargon. It is the hinge the whole claim turns on.

What the policy actually says

Insurers do not describe the peril loosely. Churchill's home insurance policy booklet dated 09/25 sets it out as a closed list. Under "Water or oil escape": "We only cover water or oil escaping from: Fixed water or heating systems. Underground drains and pipes. Domestic appliances. Storage tanks."

Read the list again and notice what is missing. No line for rain. No line for a roof, a gutter, a flashing or a window. Rainwater has no home in that peril at all. It has to find cover elsewhere, and the only other door is "Storm or flood", a separate peril with a separate test.

The same booklet is blunt about what it will not do inside that peril. It excludes the pipe, tank or appliance itself, and "Damage caused by failure of, wear and tear to, or lack of, grouting or sealant." Its plain-English page says it again: "We will deal with the resultant damage of a water leak... We will not cover the source of the leak unless you have Home Emergency cover on your policy."

So the peril pays for the damage the escaped water did. It does not pay for the pipe. That is why trace and access cover exists as a separate line with its own limit, £2,500, £5,000 or £10,000 in that booklet depending on tier. It pays for "removing and replacing any part of the buildings to find the source of escaping water or oil". Escaping. From tanks, pipes, appliances or fixed heating systems. If the water came in from outside, trace and access has nothing to attach to.

The storm test is far harder than people expect

Homeowners assume that if rain got in, storm cover applies. It rarely works out that way.

The Churchill glossary defines a storm as violent weather with minimum wind speeds of 48 knots or 55 mph, or torrential rainfall of at least 25mm per hour, or snow to a depth of at least 30cm in 24 hours, or hail intense enough to damage hard surfaces. It then adds the sentence that does most of the work in declined claims: "Well-maintained homes are unlikely to be damaged by weather less violent than a storm."

Where a policy has no numeric definition, the Financial Ombudsman Service applies three questions to a refused storm claim: was there a storm on the date the damage happened, is the damage consistent with what it generally sees as storm damage, and were storm conditions the main cause rather than a factor that meant the damage might have happened anyway. If the answer to any one is no, the ombudsman will normally say the insurer was entitled to decline.

Published decision DRN-5778796 shows how that plays out. A lounge ceiling came down in September 2024 and the insurer declined the whole claim on the basis there was no insured event. The ombudsman found a peak gust of under 30mph but rainfall of 12.6mm per hour on 22 September 2024, and accepted that as a rainstorm. The claim still failed, at the second and third questions, because photographs showed previous repairs to the main roof, missing mortar where the flat roof met the brick wall, and guttering already in poor condition. The conclusion: "the heavy rainfall simply highlighted existing issues with the property."

Now put the two documents side by side. Rainfall of 12.6mm per hour was enough to be a storm under a policy that did not define the word. Under a policy that defines a storm as 25mm per hour, the same rainfall is not a storm at all, and question one fails before anyone looks at the roof. Which policy you bought changes the answer to a question about the weather.

One more thing from that decision. The ombudsman directed the insurer to reconsider the internal damage under the remaining perils, including accidental damage and escape of water. The insurer had declined the whole claim on a storm test without asking whether a pipe was also leaking. That failure is common, and it costs claimants months.

Who has to prove what

This is where "matter of fact rather than opinion" stops being a slogan.

The Financial Ombudsman Service is explicit about the burden. On an insured event question: "it's down to you, as the policyholder, to show the damage was likely caused by an insured event." It says it will look at all the evidence, but particularly "what events the policy covers" and "professional reports and photos about the damage".

The burden then flips. If the insurer accepts the damage looks like an insured event but declines using an exclusion, such as a gradually operating clause, "the insurer must prove that it's fair and reasonable for it to rely on this policy term to decline the claim". And if it cannot show the damage happened gradually, the ombudsman is likely to tell it to pay.

Two consequences follow. Nobody wins a water damage argument with an adjective; "it looks like a leak" is not evidence, from either side. And the evidence that does decide it is technical: reports, photographs, test results, dates. That is the whole reason a leak detection survey has value in a claim beyond fixing the pipe.

How escape of water is proved

Every one of these tests returns a yes or a no. None of them requires an opinion.

• Static meter test. Close every outlet in the property. Read the meter, wait, read it again. If the dial has moved with nothing drawing water, water is leaving the pressurised system. Then close the internal stopcock and repeat: movement now points at the underground supply pipe rather than the internal pipework. It is the one test a homeowner can run before anyone attends.
• Sealed heating system pressure test. A sealed system that needs topping up is losing water somewhere. Isolate the boiler at the flow and return, pressurise the circuit and watch the gauge. If the circuit holds and the boiler does not, the loss is in the appliance. If the circuit drops, the loss is in the pipework, and on a system in a screed that is usually under the floor.
• Tracer gas. The pipe is drained and charged with a hydrogen and nitrogen mix. The gas is smaller and lighter than water and finds the same defect, then rises through screed, subfloor and tiles to a surface detector. Gas can only come out of a hole. When it surfaces in one repeatable spot, that is the hole.
• Acoustic correlation. Pressurised water forced through a small orifice makes noise in a predictable frequency band. Ground microphones locate the source along a pipe run by comparing arrival times at two points.
• Thermal imaging and moisture mapping. A hot water leak produces a warm plume with a diffuse edge that grows over minutes. A cold feed does the opposite. Moisture meters give a gradient, and the wettest point is rarely the visible point.

These get used together rather than singly, because each one rules something out. The detail of each method matters less than the sequence: least invasive test first, and stop as soon as the answer is unambiguous.

How ingress is proved

Just as concrete, but on a different clock.

• Correlation with weather. Ingress tracks rain. It appears during or shortly after rainfall and stops in dry weather. A pipe leak does not care what the sky is doing. Dating each appearance is real evidence, because the insurer and the ombudsman both pull weather records for the address.
• Directionality. Wind-driven rain hits one elevation. If damp appears on the south-west face after a south-westerly and never after a northerly, that is the envelope, not a pipe.
• Water trace testing. A hose is run on one zone of the elevation at a time, starting at the lowest and working up, twenty minutes or so per zone, with someone watching inside. It has to be bottom-up. Wet the top first and everything below it is contaminated for the rest of the day.
• Water chemistry and staining. Mains cold water is clean and chlorinated. Central heating water is dosed with corrosion inhibitor and carries black iron oxide, so a heating leak stains dark. Water that has travelled through masonry carries salts out of the brick with it and dries to a crystalline tide line. The stain itself tells you something before any instrument leaves the van.
• Roof and gutter inspection. Ponding, splits at upstands, perished lead flashing, a gutter joint that has dripped down a wall for two winters. Flat roofs fail at the detail, not in the field.

There is a third answer that is neither. Condensation produces damp with no source at all, and has its own code of practice, BS 5250:2021, Management of moisture in buildings. If a wall is cold, unventilated and wet in winter only, a plumber pulling up floorboards will find nothing, correctly. Moisture mapping separates the three cases because it measures rather than guesses.

What our own surveys show

Across our last 400 leak detection surveys, the cause of the leak was found in 95% of cases. In 2% the water was coming from a neighbouring property. In 3% it was not a plumbing leak at all: gutters, roofs and other structural defects letting water in.

Three per cent is a small number and we are not going to inflate it. But it is roughly one survey in thirty-three, and those are the households whose escape of water claim was never going to be paid, however long it ran. Finding that out on day one beats finding it out in month three, after two adjuster visits and a declined claim.

The 2% is its own trap. In a flat, water arriving through your ceiling from above is still escape of water, but it is somebody else's escape of water, and it becomes a question of whose policy and whose liability rather than a question of plumbing. We cover London out to the M25, across all 32 boroughs, and in the inner boroughs that scenario is a large share of what we attend.

The cases that genuinely sit in the middle

Some situations are neither, and the honest answer is that they fall out of cover altogether.

A leaking shower is the commonest. Water leaving the shower head is doing what it was built to do. If it then passes through failed grout or perished sealant into the floor below, no closed system has failed, and the Churchill wording excludes damage caused by failure of, wear and tear to, or lack of grouting or sealant in plain terms. That booklet also excludes overflow from taps left on in sinks, basins, bidets, showers and baths unless full accidental damage cover has been bought.

The other awkward case is both at once. A roof that has leaked for two winters and a live pipe leak of six weeks can share a ceiling. Insurers apportion, and a survey that reports only one of the two gets the claim reopened later. We would rather write both into the report on the day.

What to do in the first week

Photograph everything, with dates. Note whether the damp appeared during rain or independently of it. Find your policy wording, not the marketing page, read the peril list for "escape of water" or "water or oil escape", and find your trace and access limit.

Report it before you spend. The ombudsman notes that the usual practice, in line with policy terms, is to contact your insurer first, and that the cost of your own professional advice may not be covered, especially if what happened turns out not to be an insured event. Our insurance leak claims page sets out how the survey report is normally used.

Then get the survey done. Standard leak detection is charged at a fixed hourly rate of £160, and a typical survey runs two to four hours, so £320 to £640 in total. Heating, mains and trace and access work is £200 an hour, moisture mapping from £175 an hour, emergency attendance from £220 an hour. The rate is agreed before we travel and does not rise on the day. Behind it sits no find, no fee: if we attend a confirmed live leak and cannot locate it, the detection time is not charged. Full detail is on our pricing page.

If you have a stain, a pressure drop or a meter that will not sit still, call 020 7123 8560. We will tell you which tests your situation needs and roughly how long they take, and we will say so if we think the answer is a roof rather than a pipe.