---
title: "What inspectors look for in your Legionella records"
source_url: https://legionella.io/articles/what-inspectors-look-for-in-your-legionella-records/
canonical_url: https://legionella.io/articles/what-inspectors-look-for-in-your-legionella-records/
pillar: "Digital Logbooks & Record Keeping"
summary: "How an inspector reads your Legionella logbook: the records that survive scrutiny, the gaps that quietly fail an audit, and how to test your own evidence."
primary_keyword: "Legionella inspection records"
date_published: 2025-10-06
date_reviewed: 2026-06-26
author: "Legionella.io editorial team (REMOTE TECH LTD)"
reviewed_against: "HSE L8 and HSG274 guidance"
region: "United Kingdom"
license: "(c) REMOTE TECH LTD. Quote freely with attribution and a link to source_url."
---

# What inspectors look for in your Legionella records

An inspector almost never reads your logbook cover to cover. They pull a thread. They pick one outlet, one month, or a single out-of-range temperature, and they follow it: who recorded it, what they did next, who signed it off, whether the fault ever actually closed. If the thread leads somewhere, the rest of the file is presumed sound. If it dead-ends, every other tidy page becomes suspect.

So the records that survive scrutiny are not the neatest ones. They are the ones where every loose end goes somewhere. That is the lens to read your own Legionella inspection records through, and it is a different lens from "have we filled in the sheet".

## How an inspector actually reads the file

Depending on the premises, the visit might come from HSE or from your local authority's environmental health team [3]. Either way, the method is similar, and it is rarely a line-by-line audit. The inspector cross-checks three things against each other: the paperwork, the physical system, and the person standing in front of them.

That triangulation is where weak records come apart. If the log says a sentinel tap hit 55°C last Tuesday, but the calorifier has been flagged faulty for a month and the responsible person can't explain the limit, the mismatch is what gets written up. The inspector is testing whether your records describe a control scheme that is actually running. L8 frames the duty as keeping records of the precautions and monitoring and appointing competent people to manage the system [1]; HSG274 sets out the monitoring that sits behind those records [2].

## The records an inspector will ask to see

Read this as a dry run. Work through it the way they will, and note anything you couldn't produce in five minutes.

**The foundation — does a control scheme exist on paper?**
- Produce the current Legionella risk assessment, with its date, the assessor's name, and evidence they were competent to do it.
- Show the written scheme of control: the system described, the control measures, who does what, and the limits that define "in range".
- Name the responsible person in writing, and be able to put them in the room.
- Keep an asset register that matches the building as it is today, not as it was at handover.

**The routine evidence — is the scheme running?**
- Log temperature monitoring at sentinel and representative outlets, with readings, dates and the initials of whoever took them.
- Record flushing of infrequently used outlets, tied to the specific outlet rather than a vague "all low-use taps".
- Keep tank inspections, TMV servicing, cleaning and descaling filed against the asset they relate to.
- File sampling reports where the risk assessment calls for them, with method and location stated, not just a pass/fail line.

**The exception trail — what happened when something went wrong?**
- Show every out-of-range reading, not just the ones that resolved easily.
- Show the action taken for each one, with a date.
- Show the close-out: the recheck that confirms the fault is fixed, signed.
- Evidence the escalation route — the point at which a result lands on the responsible person's desk.

**The management layer — is anyone actually steering?**
- Show the risk assessment has been reviewed at suitable intervals and after changes to the system or its use.
- Hold training and competence records for everyone carrying out tasks, and evidence of your contractor's standing, such as membership of a recognised service-provider scheme [5].
- Provide some sign of review: a person reading the data and acting on trends, not merely collecting sheets.

## Read your own file before they do

Pick a random recent month and choose the single worst reading in it. Then follow that one reading to its close-out, exactly as an inspector would. If you can't get from "47°C at the sentinel tap" to "rechecked at 56°C, signed, ten days later" without phoning three people, you've found your gap before someone with enforcement powers does.

The useful habit is to record the decision, not just the task. "This outlet is flushed weekly because use is intermittent; a missed flush escalates to the responsible person; repeated misses trigger a use-pattern review" tells an inspector that a thinking person owns the control. A bare tick tells them nothing about whether anyone understands why the task exists.

## What too-clean records tell an inspector

Here is the part most guidance skips. A logbook where every temperature for two years sits perfectly in range does not reassure an experienced inspector — it makes them suspicious. Real hot and cold systems wobble. A cold tap creeps up in a heatwave, a long run loses a degree, a calorifier needs a nudge after a cold snap. A record that shows those wobbles being caught and corrected is far stronger evidence of a live control scheme than an unblemished sheet that looks transcribed at month-end.

So out-of-range readings with documented responses are the proof, not the embarrassment. The thing to be ashamed of is the perfect-looking column, or the anomaly that simply vanishes from the record with no fix attached.

This is also where digital logbooks both help and mislead. A green dashboard is a summary, not evidence, and an inspector will ask to see the underlying entries behind the green. A well-built system timestamps the entry, ties it to the asset and the person, and makes the exception trail impossible to misplace — but the same averaging that produces a tidy dashboard can also bury a run of failures. If you are choosing how to store all this, [Analysing digital logs for performance improvements](https://legionella.io/articles/analysing-digital-logs-for-performance-improvements/) covers reading the trends rather than the headline.

## The gaps that quietly fail an audit

Most failed inspections are not caused by one dramatic error. They are caused by small, recurring omissions that, stacked together, suggest no one is really in control:

- **Gaps in the record.** A missing month reads as "not done", not "lost". Silence is the worst answer.
- **Orphan readings.** A number with no asset, no units and no initials proves nothing and protects no one.
- **A drifted asset register.** New outlets that never made it onto the list, or removed ones still being "monitored" on paper.
- **Open actions that never close.** A remedial action raised, then abandoned, is worse than one never raised.
- **Contractor work that never joins your trail.** Sampling results or servicing that live only in the contractor's system are evidence you cannot produce on the day.

## A note on judgement and limits

Inspectors apply judgement, and what satisfies one visit on one site is not a fixed national tick-list. The detail of what is expected flows from L8, HSG274 and, above all, your own risk assessment, which sets the limits and frequencies for your particular system [1][2][4]. Treat the checklist above as a way to pressure-test your evidence, not as a substitute for advice from a competent person who knows your building. None of this is legal guidance.

## FAQ

### How far back do my records need to go?
L8 sets out how long different records should be retained, and the period varies by record type — a figure commonly quoted for monitoring results is around five years, with the risk assessment kept current and held for a period after it is superseded [1]. Confirm the exact periods against L8 for each record you hold. Whatever the number, an inspector reads a record that stops eighteen months ago as a control scheme that stopped eighteen months ago.

### Will a digital logbook satisfy an inspector, or do they want paper and wet signatures?
Format is not the test; traceability is. A digital system that shows who did the task, when, against which asset, what the result was and what happened next is generally stronger than a paper sheet — provided you can produce the underlying entries and not just a dashboard [1]. The failure mode is a slick summary that no longer ties back to who actually did what.

### What happens if an inspector finds a gap?
It depends on the gap and what it implies about real control. Outcomes range from verbal advice, to a written notice requiring improvement, to firmer enforcement where exposure risk looks serious [3]. The weakest position is a gap you cannot explain. A gap with a documented reason and a fix already in progress is a very different conversation.

## Related reading

- [Pre-flush vs post-flush sampling: which to use when](https://legionella.io/articles/pre-flush-vs-post-flush-sampling-which-to-use-when/)
- [Seasonal buildings: managing intermittently used properties](https://legionella.io/articles/seasonal-buildings-managing-intermittently-used-properties/)
- [Cloud vs on-premise: where to host your Legionella data](https://legionella.io/articles/cloud-vs-on-premise-where-to-host-your-legionella-data/)
- [Analysing digital logs for performance improvements](https://legionella.io/articles/analysing-digital-logs-for-performance-improvements/)

## Next step

Before your next audit, block out twenty minutes and run the thread test on last month's records: pick the worst reading, follow it to a signed close-out, and fix whatever you couldn't trace. Do that every month and the real inspection becomes a formality instead of a fire drill.

## 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, "Legionnaires' disease - what you must do". https://www.hse.gov.uk/legionnaires/what-you-must-do/index.htm
[4] HSE, "Testing and monitoring your water system for legionella". https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm
[5] Legionella Control Association, "Code of Conduct for Service Providers". https://www.legionellacontrol.org.uk/