My meter moves with everything off. Why can that not tell me where the leak is?

Because it is a count of volume rather than a description of a path. The meter registers water passing it and the water then goes wherever the pipe goes. Two properties with identical readings can have leaks fifty metres apart. That is not a shortcoming of your meter, it is what measuring volume at a point means.

My meter is not moving. Does that prove there is no leak?

No. A domestic meter has no specified accuracy below a stated low flow, and a typical one does not begin to turn at all until the flow reaches a few tenths of a litre an hour. So a still meter rules out anything above that threshold and says nothing at all below it. It is a negative with a floor under it rather than a clean zero.

Could the leak be somewhere the meter never sees?

Yes, and it is worth ruling out before spending anything. A sealed heating system topped up by hand, a swimming pool, and anything fed from a private supply or a rainwater system all lose water that never crosses the meter. So does a leak on the far side of a shared supply, depending on where your meter sits.

I did the test once and it was clean. Is that enough?

It is one sample of one night. A loss that only happens when a particular valve is under pressure, or when a pump runs, or when the ground is dry, will pass a single test and fail another. If the bill still says otherwise, repeat it at a different time before concluding anything.

What is the most useful thing I can do without paying anyone?

The stop tap test, done carefully. Close the internal stop tap, leave everything off, read the meter twice over a measured interval. If it still moves, the loss is on the underground supply pipe. If it stops, it is inside. That halves the search area, and it is the last narrowing that costs you nothing.

When a meter test cannot settle it

A water meter counts volume passing one point over a period. There is nothing in that quantity that could tell you where the water went, so no meter test of any quality gives a position. What a meter test can leave open is wider than that though: a loss too slow for the meter to register, a loss that only happens sometimes, and a loss that never crosses the meter at all are all cases where the reading is not the answer. Closing the internal stop tap and repeating the test is the last narrowing available to you for nothing.

Last reviewed 14 August 2026

The short answer

A meter measures how much water passes one point in a period. No amount of that measurement adds up to a position, because position was never one of the things being measured.

That is worth stating as arithmetic rather than as an apology. A better meter, a longer test or a more careful reading would all give you a better rate. None of them would give you a location, because there is no location in the quantity.

How this works in practice

Five patterns leave a meter test unsettled, and they are not all the same problem. Three are limits of the measurement, one is a limit of the sample, and one means the meter was never going to see it.

The pattern	What the meter tells you	What it leaves open
It moves steadily with everything off.	There is a real, continuous loss, and how fast it is.	Where. Nothing about a volume contains a position.
It does not move at all.	There is nothing above the meter's low flow threshold.	Anything below it. A typical domestic meter does not begin to turn until the flow reaches a few tenths of a litre an hour.
It moves, but only sometimes.	That one test caught, or missed, a loss.	Whether it is intermittent or whether you sampled the wrong night. One test is one sample.
The bill is high and the meter is clean.	Nothing is leaking past the meter now.	Whether it leaked before, and whether the extra volume was use rather than loss.
Water is going somewhere off-meter.	Nothing, correctly.	A heating circuit topped up by hand, a pool, a private supply or a rainwater system all lose water the meter never counts.

What a meter test cannot settle

The second row is the one most often read as good news. A still meter is a negative with a floor under it: it rules out everything above the threshold and says nothing whatever below it. On a slow loss that has been running for years, below the threshold is exactly where it is.

A loss too slow to register is a small volume that is still going somewhere, and the honest reading of that case is set out separately. quantify the loss at the meter first also covers how far down the scale a domestic meter is specified to be accurate at all.

What changes the answer

Three things you can still do for nothing, in the order that narrows the most.

Close the internal stop tap and repeat the test. Still moving means the underground supply pipe. Stopped means inside. That is one bit of information and it halves the area.
Repeat the test at a different time, and on a different day of the week. That is the only way to separate an intermittent loss from a night you happened to catch.
Account for everything fed off-meter before assuming the meter is wrong. A heating system you have topped up, a pool, an outside supply on a separate feed, or a shared arrangement with a neighbour.

The stop tap test deserves care disproportionate to how simple it is, because it is the last narrowing available to you at no cost. Everything after it is a paid search over whatever area you hand over, and the area you hand over is decided here. Getting it right is the difference between paying to search a garden and paying to search a house and a garden.

It is worth doing over a long enough interval to be readable, for the same reason the original test is. If your loss is near the bottom of the scale, an hour with the stop tap closed may move the meter by less than one increment, and you will have learned nothing rather than having learned there is no loss on the supply pipe. Give it the same overnight interval you gave the first test, at several hours rather than one.

Assumptions and sources

Aquadis+ DN15 and DN20 specification sheet

Itron · manufacturer source

Q1 minimum flow rate, DN15, Q3 2.5, R160: 15.6 l/h
Q2 transitional flow rate, DN15, Q3 2.5, R160: 25 l/h
Q3 permanent flow rate, DN15: 2500 l/h
Typical starting flow rate, DN15: 0.4 l/h
Typical starting flow rate, DN20: 2 l/h
Q1 minimum flow rate, DN20, Q3 4.0, R160: 25 l/h
Flow rate at which accuracy is within +/- 5%, DN15: 3 l/h
Flow rate at which accuracy is within +/- 2%, DN15: 5 l/h
Minimum scale interval, DN15 and DN20: 0.02 l

Litres in a cubic metre

Exact by definition

SI definition. The litre is defined as one cubic decimetre, so one cubic metre is exactly 1,000 litres.

Hours in a day

Exact by definition

Definition of the day as 24 hours.

The meter thresholds referenced here are the ones used throughout this site: a domestic meter is specified for accuracy only above a stated low flow, and a typical unit does not begin to register until somewhat above zero. Both are named in full with their sources on the water meter leak test page, and they are the reason a still meter is treated as a floor rather than a zero.

Nothing on this page claims how common any of these patterns is. No source was found for how often a household loss is intermittent, how often it sits below meter resolution, or how often it turns out to be off-meter entirely. They are listed because each one is possible and each one changes what your reading means, not because any of them is likely.

The boundary itself needs no citation. That a measurement of volume at a point does not contain a position is not an empirical finding about meters, it is what the quantity is. Every other limit on this page is a fact about instruments; that one is a fact about arithmetic.

What to do next

Finish the free work before deciding anything, and write down what each step established.

Repeat the meter test with the internal stop tap closed, over the same long interval, and record both readings and the times.
Repeat the original test on a different night if the first was clean and the bill still disagrees.
List everything in the property that uses water without crossing the meter, and rule each one out.
Keep the readings, the dates and the interval together. They are what turns a suspicion into a measured rate, and a measured rate is the thing every later step is built on.

At the end of that you know three things: how fast you are losing water, subject to the meter's floor; which side of the internal stop tap it is on; and whether it is continuous or comes and goes. That is a genuinely useful position to be in and most people never get to it.

You still do not know where along the pipe, at what depth, or whether it is one leak or several. Those are the questions a meter cannot answer at any level of care, because none of them is a volume. That is the boundary of what this site can work out with you, and saying where it is seems more useful than pretending it is somewhere else.

Common questions

My meter moves with everything off. Why can that not tell me where the leak is?: Because it is a count of volume rather than a description of a path. The meter registers water passing it and the water then goes wherever the pipe goes. Two properties with identical readings can have leaks fifty metres apart. That is not a shortcoming of your meter, it is what measuring volume at a point means.
My meter is not moving. Does that prove there is no leak?: No. A domestic meter has no specified accuracy below a stated low flow, and a typical one does not begin to turn at all until the flow reaches a few tenths of a litre an hour. So a still meter rules out anything above that threshold and says nothing at all below it. It is a negative with a floor under it rather than a clean zero.
Could the leak be somewhere the meter never sees?: Yes, and it is worth ruling out before spending anything. A sealed heating system topped up by hand, a swimming pool, and anything fed from a private supply or a rainwater system all lose water that never crosses the meter. So does a leak on the far side of a shared supply, depending on where your meter sits.
I did the test once and it was clean. Is that enough?: It is one sample of one night. A loss that only happens when a particular valve is under pressure, or when a pump runs, or when the ground is dry, will pass a single test and fail another. If the bill still says otherwise, repeat it at a different time before concluding anything.
What is the most useful thing I can do without paying anyone?: The stop tap test, done carefully. Close the internal stop tap, leave everything off, read the meter twice over a measured interval. If it still moves, the loss is on the underground supply pipe. If it stops, it is inside. That halves the search area, and it is the last narrowing that costs you nothing.