Fire Drill Record Sheet template

2. Drill details — complete before the alarm sounds

• Premises: [site name and full address]
• Responsible person: [name, role]
• Drill conducted by: [name, role]
• Date: [date] | Time alarm activated: [hh:mm]
• Announced or unannounced: [unannounced / announced to all / announced to managers only]
• Call point, detector or zone used to initiate: [location — rotate this between drills so you are not always testing the same route]
• Number of people on site at activation: [staff] staff, [contractors] contractors, [visitors] visitors
• Shift or session covered: [e.g. Tuesday late shift, weekend housekeeping, night porter]
• Weather and conditions: [dry / heavy rain / dark / icy underfoot] — record it, because it changes how people behave at the assembly point
• Complication introduced, if any: [none / exit [x] taped off as blocked / marshal [name] absent / lift out of service]
• What this drill is testing: [one specific question — e.g. whether the kitchen can isolate gas and clear in under two minutes, or whether the first-floor personal emergency evacuation plan (PEEP) buddy arrangement works]
• Alarm receiving centre, neighbours and [any at-risk operation] notified in advance: [yes/no — by whom]

3. Drill record: exits, timings and roll call

One row per evacuation point or exit. "Time to clear" is the interval between alarm activation and the last person leaving by that route, taken by the observer standing at it. "Roll call complete" means the people who used that route have been matched against the register at the assembly point — not simply that somebody saw them outside.

Write the issue in the row where it happened, not in a general comments box at the bottom. A fault attached to a specific door gets fixed; a fault attached to "the building" does not.

4. Observers and what they were asked to watch

Observers are the reason a drill produces findings rather than a number. Post one at each exit and one at the assembly point, brief each on a single thing to watch, and tell them not to intervene unless someone is at risk — an observer who starts marshalling stops observing.

Observers are not exempt from the evacuation. Give them a position on an escape route or outside it, never inside a zone they would have to be swept out of.

5. Roll call: accounting for everyone

Timing the evacuation is the easiest number to collect and the least useful one. What matters in a real fire is whether the roll-call lead can tell the incident commander, within a minute of their arrival, exactly who is unaccounted for and where they were last seen. Record the arithmetic, not just the clock.

• Number on site at activation, from the staff register and the signing-in book: [number]
• Number accounted for at the assembly point: [number]
• Time the last person reached the assembly point: [hh:mm]
• Time all persons were confirmed accounted for: [hh:mm]
• Unaccounted for at the close of roll call: [names / none] — last known location: [location]
• How visitors and contractors were accounted for: [signing-in book / host responsible / badge system] — did it work? [yes/no and why]
• Anyone who did not evacuate, and why: [name — did not hear the alarm in [area] / continued a task / used a lift / refused]
• PEEPs exercised this drill: [names / none] — outcome: [worked as written / needs revision — detail]
• Anyone who re-entered before the all-clear was given: [names / none]
• All-clear given by [name, role] at [hh:mm]

6. Debrief — findings and changes

Hold the debrief at the assembly point within [15] minutes of the all-clear, while people are still standing where the problems happened, and write the answers straight onto this sheet. A debrief postponed to the next management meeting produces a tidier document and a shorter list.

• What worked and should be kept: [e.g. the second-floor sweep finished inside a minute; the kitchen isolation sequence was followed]
• What went wrong: [each failure plainly — obstructed route, wedged door, alarm not audible in [area], [room] not swept, marshal absent with no deputy]
• Alarm audibility and detection: [any area where the alarm was not clearly heard — [locations]; reported to [name] on [date]]
• Fire doors, escape routes and emergency lighting: [any door that stuck or was propped, route obstructed, or light out — [details]; defect reported to [name] on [date]]
• Changes to the fire evacuation procedure: [what changes, who edits it, by when — or "none required"]
• Changes to training or briefing: [who needs retraining, on what, by when — e.g. all new starters since [date] to walk the routes again]
• Changes to the fire risk assessment: [what the drill showed that the assessment did not anticipate — refer to [name] for review by [date]]
• Actions raised: [action] — owner [name] — due [date]. Repeat for each; copy all of them into the health and safety action log.
• Date of the next drill: [date] — covering [the shift, area or scenario this drill did not test]

7. Sign-off

• Drill conducted by: [name, role] — signature: [__________] — date: [date]
• Roll-call lead: [name, role] — signature: [__________] — date: [date]
• Responsible person review and sign-off: [name, role] — signature: [__________] — date: [date]
• Findings communicated to staff by [team briefing / noticeboard / email] on [date]
• Actions transferred to the health and safety action log held at [system/location]: [yes/no] — by [name]
• Fire risk assessment review triggered by this drill: [yes — referred to [name] on [date] / no]

8. What people look for in a drill record

Inspectors, insurers and incoming responsible persons tend to ask for the same things, and none of them is the evacuation time.

• A series, not a sheet. One drill record with no predecessors suggests the drill was arranged for the visit.
• Coverage of every shift pattern across the year, so the night team and weekend staff have actually experienced an evacuation rather than read about one.
• Findings. A run of records saying "no issues" in a building with six exits and a kitchen reads as a drill nobody observed.
• Closed actions. An issue raised at three consecutive drills with no completion date is the most damaging thing a record can show.
• Traceability into the documents: a finding here should be visible as an edit to the fire evacuation procedure, the training record, or the fire risk assessment.
• Evidence that people needing assistance were included — PEEPs exercised, not just filed.
• Consistency with the recorded fire risk assessment findings that article 9 requires, including any group identified as especially at risk.

9. How to run a drill that tests the plan, not the alarm

1. 1Decide the question before you set the date. "Does the alarm work?" is a routine call-point test, not a drill. "Can the late shift clear the first floor without the lift?" is a drill.
2. 2Vary the initiation point, the blocked route and the time of day between drills. Always sounding the same call point at 10am on a Tuesday tests one route and one set of people.
3. 3Brief the observers, not the participants — but do warn anyone for whom a surprise evacuation is genuinely unsafe or unworkable [named roles — e.g. staff mid-treatment, drivers, lone workers], and tell your alarm receiving centre and neighbours before you sound the alarm.
4. 4Let it run properly. Stopping the drill when the last person leaves misses the part that matters: the roll call, the visitor arithmetic and the marshal reports.
5. 5Cover the awkward hours across the year — night shift, weekends, and the part-time staff who are never on site on a Tuesday morning.
6. 6Time the evacuation, but judge the drill on the roll call, the findings and whether the actions get closed before the next one.

10. Records, retention and review

Keep completed drill sheets with the fire risk assessment, the alarm test log and fire training records in [system/location], and retain them for [retention period — set this in your data retention policy]. The Order's recording duty bites on the risk assessment; the drill sheets are how you show its arrangements are live rather than theoretical.

Sheets that name individuals are personal data, and PEEP details can include information about a person's health, which carries stronger protection — keep those securely and share them only with the people who need them to get someone out of the building.

Review the drill programme [annually], and immediately after any drill that finds a fault in the escape routes, the alarm, the roll call or a PEEP. Owner: [name/role]. Next review due: [date].