An inspector standing in a building services cupboard with your emergency lighting log has a limited amount of time and a fairly short list of things they want to see. They are not reading it cover to cover. They are checking a handful of specific points, and if those points hold up the rest of the file is usually taken on trust.
Knowing which points those are is the difference between a five-minute look and a follow-up visit. This guide walks through what gets read first, in roughly the order it gets read.
If you maintain these records, build them in the emergency lighting test register. For the field-level detail, see the emergency lighting test register and AS 2293.2 guide, and for the arranging side, the six-monthly checklist for facility managers. Browse more electrical forms for the rest of your building records.
The dates get read first
Before anyone looks at a single fitting, they look at the spacing of the services.
Emergency and exit lighting should be inspected and electrically re-tested every six months under AS/NZS 2293.2. A log showing services in March and September for three years running tells its own story. A log showing March, September, then nothing until the following August, tells a different one.
Two patterns draw attention. The first is a gap, where a cycle was skipped and the building went twelve months on one service. The second is bunching, where three services appear within a few weeks of each other because someone realised the records were behind and caught up on paper. Bunched dates are worse than a visible gap, because they suggest the record was written to look right rather than to reflect what happened.
If a cycle genuinely slipped, the honest entry is the better one. A recorded late service with a reason is a manageable finding. A backfilled service that never happened is a different category of problem.
Does the fitting count match the building?
This is the second check and it is done by walking, not reading.
An inspector who counts eleven fittings on a floor and finds nine on the register has a reason to doubt everything else in the file. It is a quick test and it is decisive, which is exactly why it gets used.
The count moves when buildings change. A fitout that adds a corridor adds fittings. A tenancy reconfiguration removes them. A register reporting an unchanged count in a building that has visibly been altered is being copied forward from the last service rather than rebuilt from the site.
The fix is that the register is built against a current schedule, and that new or removed fittings are reflected the cycle they change. If a fitting was removed, saying so on the record is better than it silently disappearing between two services.
Every fitting, not a summary
The most common weak log is a certificate rather than a list. One page, a building name, a date, and a line reading "emergency lighting tested and found compliant".
That is a claim. The record an inspector wants is the evidence behind it: each fitting with an identity, a test date, and its own result. A log that lists only the failures is a variant of the same problem, because there is no evidence the rest were tested at all and no baseline to compare against next cycle.
Identity matters as much as the result. "Level 2 corridor, east end, above door 2.14" can be found by someone who has never been in the building. "Corridor light" cannot, and an unfindable fitting is one that cannot be rectified or re-checked.
Did the discharge test actually run?
This is where an experienced inspector spends their attention, because it is the check most easily skipped.
Switching a fitting on and seeing the lamp work proves the lamp. The discharge test interrupts the normal supply so the fitting runs on its own battery, and the fitting only passes if it stays lit at the required level for the full rated duration. WorkSafe Victoria describes that as maintaining the required level for at least one hour after the loss of normal lighting. The battery is the part that degrades, so the discharge test is the part that means anything.
The tell is timing. If a log shows a hundred fittings tested across a two-hour visit, the durations do not fit. Each fitting has to run for its rated duration, and while fittings can be discharged in parallel, the numbers still have to be plausible against the time on site.
A log that records results in a way that shows the test ran, rather than just a pass tick, answers the question before it is asked.
Failures with no action
If there is one finding that appears more than any other, it is this.
A fitting marked as failed with nothing recorded against it leaves an open safety item. The inspector does not know whether it was replaced that afternoon, quoted and forgotten, or is still sitting dark above a stairwell eight months later. Neither, usually, does the building owner.
The entry that closes it is short. What failed, what was done, and when.
Where rectification is a separate visit, the rectification record should be filed alongside the register that reported the failure. An inspector following one failed fitting from the report to its closure and finding both documents together is the moment a well-kept file earns its keep.
Who signed it
The last check is the simplest. The record should name the competent person who performed the routine service, and the business behind them.
Records of testing should be kept on site and made available for auditing, and a record nobody is named on is hard to rely on. It also makes practical follow-up impossible, because the obvious next step after a question about a result is to ask the person who took it.
What a clean log looks like
Pulled together, the log that passes without much discussion has:
- Services spaced on the six-monthly cycle, with any slip visible and explained rather than smoothed over
- A fitting count that matches the building, with additions and removals reflected when they happened
- Every fitting listed with a findable location, a date, and its own result
- Evidence the discharge test ran to the rated duration, not just a lamp check
- An action against every failure, with rectification records filed alongside
- The tester and business named
None of that is exotic. It is what the record looks like when it is written at the fitting rather than reconstructed at a desk.
Where the form helps
The emergency lighting test register produces this shape by default, because it asks for the per-fitting detail as you go rather than leaving it to whoever is writing up the visit. Each fitting is entered with its identity, result, and dates, and a failure prompts for the action before the record can be finished, which is what stops the open-failure finding.
Business, tester, and licence details are saved and reused across jobs, so the naming question is answered every time without retyping. The form flags missing fields before download, you preview the entries on the register layout, then download the finished PDF to hand over and store with the building records.
Tradie Forms maps your entries onto that layout. The competent person performing the service remains responsible for the testing, the results, and checking the downloaded PDF before it becomes the record.
Official references
Check AS/NZS 2293.2:2019 at the Standards Australia store for the routine service and maintenance procedures, National Construction Code Part E4 for how emergency lighting and exit signs are required, WorkSafe Victoria on emergency lighting for the six-monthly test and record-keeping expectations, and the Queensland fire safety inspection and maintenance guidelines for how the service and its records are treated. Inspection regimes and retention expectations vary by state and building classification, so confirm what applies to the building you maintain.
