Ask ten water-safety contractors how often you should flush and test, and you’ll get the same answer phrased ten ways: it depends. That isn’t a dodge. Frequency is something your risk assessment produces, not a number you pick first and justify afterwards. Two buildings on the same street can sit on completely different schedules and both be right.

So the question that actually helps isn’t “what’s the standard interval?” It’s “what does my system, and the way people use it, demand?” Get that reasoning straight and the calendar more or less writes itself.

Control and verification run on different clocks

First, separate two things that get routinely blurred. Flushing, temperature checks, cleaning and descaling are control monitoring — recurring proof that your control regime is doing its job. Sampling water for Legionella is verification — evidence about microbiological conditions at one point in time. They answer different questions, so they keep different schedules, and you should never swap one for the other.

HSE is explicit that the frequency and type of testing depend on the system and the risk assessment [3]. L8 and HSG274 are the documents your scheme should trace back to [1][2]. A clean sample is reassuring, but it is not a stand-in for hot water that is genuinely hot, cold water that is genuinely cold, and water that actually moves [4].

Why there is no universal number

Frequency tracks risk, and risk is never spread evenly across a building. A handful of drivers push a task towards a tighter interval: outlets that sit unused for days, long pipe runs and dead legs, cold water that drifts above its target in a warm plant room, hot returns that have cooled by the time they reach a far wing, and vulnerable people standing downstream of any aerosol. The reverse holds too. An outlet in constant use is, in effect, flushing itself, and stacking extra checks onto it buys very little.

That is why a copied-in schedule from another site is usually wrong in both directions at once: too much attention on the busy outlets, not enough on the forgotten ones.

Map the system before you set a single date

Before you write any frequency down, sketch the system as a diagram you could explain to a new starter. Draw it left to right as a flow.

Start at the left with the incoming main and the cold water storage tank — your source and storage. Draw lines from there through the distribution: the risers and horizontal runs that feed each floor, with the calorifier or water heater branching in for the hot side. End at the right with every outlet — taps, showers, the cleaner’s sink, the spray tap in the accessible toilet nobody books.

Now annotate it. Beside each outlet, write how often it is actually used, in days. Put a red dot on every dead leg and blind branch. Shade any length of pipe that runs through warm space. Against each node, note who checks it and what reading is acceptable.

The schedule falls out of the picture. The red dots and the high day-counts are where flushing belongs most often. The shaded warm runs are where temperature checks earn their place. The storage tank and the calorifier carry their own inspection intervals. And a node with nobody’s name against it is the real finding — an uncontrolled point hiding inside a tidy policy.

What the common intervals actually look like

With the map in front of you, the conventional starting points in HSG274 become something to apply rather than copy blindly. As a general expectation, little-used outlets are flushed roughly weekly to stop water standing [4]; sentinel taps — the nearest and furthest on each hot and cold service — get a monthly temperature check; showerheads and hoses are cleaned and descaled on a periodic cycle; and the cold water tank and the calorifier are inspected on a longer, typically annual rhythm, with the wider set of outlets rotated through over time [2]. Treat each of those as a default your assessment can tighten or relax with a reason, not as a fixed rule.

Sampling sits apart from all of it. There is no single routine sampling interval that applies to every system. BS 7592 covers how to choose representative sampling points and when routine, validation, commissioning or outbreak sampling is appropriate [5]. Plenty of well-run hot-and-cold systems that hold temperature and stay in use do not sample on a routine clock at all — they rely on temperature control as the primary defence and sample when the assessment, a control failure, or an investigation calls for it [3].

The failure modes worth naming

Three patterns cause more trouble than any single missed reading.

The blanket frequency is the first — one interval applied to every outlet because it is easier to administer. It over-services the busy taps and under-services the idle ones, which is precisely backwards.

Sampling theatre is the second: an annual sample treated as the headline proof of safety while temperatures drift and a dead leg quietly festers. A negative result from a badly controlled system is the most dangerous result there is, because it buys false confidence.

The unrecorded flush is the third. If housekeeping runs the taps but nobody logs it, you carry the cost of the task and keep none of the evidence. After an incident, “we definitely did it” is not a defence — the record is. (If your records side needs tightening, How long to keep Legionella records in the UK covers how long to keep them and why.)

When the frequency should change

Intervals are not set once and forgotten. Bring them forward after anything that shifts the risk: a wing taken out of service, a refit that adds dead legs, a change in who uses the building, a run of out-of-range temperatures, or a new contractor inheriting the scheme [1][2]. A schedule that has not moved in three years usually means nobody has looked, not that nothing has changed.

A word on where this stops

Every interval above is a starting point, not a ruling. The figures that end up on your schedule have to come from a competent, site-specific risk assessment that weighs your real pipework, temperatures, occupancy and the people exposed — and they should be reviewed by someone who can read the system, not only the spreadsheet. Where that assessment and this general guidance disagree, the assessment wins.

Where to start this week

Pull your current monitoring schedule and lay it beside the system map described above. For every line on the schedule, check you can name the outlet, the reason for its interval, the acceptable reading, and the escalation step when a result falls outside it. The first line you cannot account for is where your next half-day goes. If flushing low-use outlets is the weak point specifically, Flushing little-used outlets: best practices goes deeper on doing it properly; for the verification side, Legionella sampling 101: how and why to test your water covers how and why to sample.

FAQ

No. There is no blanket rule that every building must sample on a yearly clock. HSE ties testing frequency to the system and the risk assessment [3], and many lower-risk hot-and-cold systems rely on temperature and regular use rather than routine sampling. Your assessment decides whether you sample at all, and how often.

How often should little-used outlets be flushed?

Commonly about weekly, to stop water standing long enough for bacteria to build up [4] — but the interval belongs to your risk assessment, and an outlet left void for a fortnight may need flushing before it is brought back into service rather than waiting for its usual day. Log each flush so it counts as evidence, not just effort.

If a sample comes back clear, can we stretch out our flushing and temperature checks?

No. A sample verifies one moment; flushing and temperature checks are the control that keeps the system safe between those moments [3][4]. A clear result tells you conditions were acceptable when the sample was taken — it is not permission to ease off the routine that produced them.

Sources

[1] HSE, “Legionnaires’ disease. The control of legionella bacteria in water systems - Approved Code of Practice and guidance (L8)”. https://www.hse.gov.uk/pubns/books/l8.htm [2] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm [3] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm [4] HSE, “Hot and cold water systems”. https://www.hse.gov.uk/legionnaires/hot-and-cold.htm [5] BSI, “BS 7592:2022 - Sampling for Legionella bacteria in water systems. Code of practice”. https://knowledge.bsigroup.com/products/bs-7592-sampling-for-i-legionella-i-bacteria-in-water-systems-code-of-practice-1