The short answer
Score an emergency drill along two linked axes: measured speed (alarm-to-muster time compared with a target you set for your building) and decision quality (route choice, headcount accuracy, assistance for people who need help, and communication). A fast evacuation with a miscounted headcount is still a failure, and a slow but orderly one is not automatically a pass. Total an objective, weighted scorecard, debrief everyone promptly, and close corrective actions before the next drill.
Key takeaways
- A stopwatch alone is misleading: speed without decision quality (route choice, headcount, assistance) tells you little about real readiness.
- Define a defensible target alarm-to-muster time for your building and scenario before you time anything, not after the drill.
- Capture at least three objective timestamps: alarm-to-first-movement, last person to the assembly point, and headcount completion.
- Grade observable behaviours, not intentions: alarm comprehension, route selection, area sweeps, headcount accuracy, and help for people who need it.
- Debrief promptly with all participants, separate observations from causes, and log the verdict, timestamps, and open corrective actions.
- A critical error such as an unresolved headcount discrepancy voids the drill regardless of elapsed time.
Score speed and decisions together, not separately
A stopwatch alone paints a false picture of preparedness. A building can clear quickly while people take the wrong exit or the headcount never reconciles. Equally, a slow but disciplined evacuation often reveals genuine gaps in the plan. The dependable approach treats speed and decision quality as two axes that feed a single scorecard, so the verdict reflects both how fast people moved and how well they decided.
The evaluative logic of drill planning guidance makes this explicit: during the drill you determine whether the alert was clearly communicated and understood across the building, whether procedures and protocols were properly followed, and whether accountability was maintained — and you record evacuation times. Afterward you debrief on how long the drill took, what worked well, what needs improvement, and what response time would have been in a real event. In short, time is one of several observed facts, not the verdict itself.
- Speed answers the question: when did everyone reach the assembly point?
- Decisions answer the question: did people do the right things along the way?
- Issue a verdict only after weighing both axes together.
Set a defensible target time before you time anything
There is no universal "evacuate in X minutes" number that fits every building; achievable time depends on footprint, floor count, occupant load, and mobility. Before the drill, define a target alarm-to-muster time for the specific scenario, based on the design assumptions in your evacuation plan and your own drill history, rather than on a round figure. Also state the acceptable tolerance so that later scoring is not arbitrary.
On the day, capture at least three objective timestamps: time from the alarm to first movement, the time the last person reaches the assembly point, and the time the headcount is complete. Record them on the spot — memory drifts within minutes. This timing discipline also supports the regulatory expectation in the United States that the evacuation procedures in your emergency action plan actually work and that employees understand their responsibilities; conducting and measuring drills is one of the clearest ways to demonstrate that.
- Document target time and tolerance and have it approved before the drill.
- Assign a dedicated timekeeper for each assembly point.
- Use identical timestamps on every drill so you can track trends.
Score decision quality, not just rule-following
A useful way to classify errors mirrors how exercise evaluations treat them. A critical error is a decision that, in a real event, would escalate harm — for example, directing people onto a blocked route or missing a person during accountability. A non-critical error slows things down but does not worsen the situation. Grading depends on which category an individual's actions fall into.
Score observable behaviours: correct interpretation of the alarm, route selection (did people switch to the alternate route when the scenario blocked the primary one?), sweeps of restrooms and utility rooms, doors closed but not locked, headcount accuracy against the roster and visitor log, assistance for anyone who needs help evacuating, and timely reporting of discrepancies. Base each rating on what an evaluator saw or recorded, never on intention.
- Alarm understood and action started per the plan.
- Correct main or alternate route chosen for the scenario.
- Rooms swept and doors closed without being locked.
- Assistance provided to people who need help or do not know the route.
- Headcount reconciled and discrepancies reported up the chain.
A weighted scorecard produces a clear verdict
To turn observations into a decision, score each block with weights: speed 40%, decision quality 40%, and coordination and documentation 20%. Within each block, rate criteria on a simple 0/1/2 scale with written anchors (not done, partly done, done). The weighted total maps to a verdict: acceptable, good, or requires repeat. Written anchors remove "gut-feel" ratings and make results reproducible across evaluators and drills.
Set critical-failure rules before the drill starts. Certain outcomes void the exercise regardless of elapsed time — for example, a headcount discrepancy that is never resolved, or people being routed into a hazardous area. This prevents the common trap in which a fast but unsafe drill is reported as successful.
- Speed 40%: alarm reaction, last person to muster, headcount completion.
- Decisions 40%: route, sweeps, assistance, reporting.
- Coordination and documentation 20%: headcount accuracy, communication, all-clear.
- A critical error overrides the point total and fails the drill.
Debrief, document, and close corrective actions
Hold the debrief promptly with everyone who took part, while recall is fresh. A structured sequence works best: restate the objectives and scenario; evaluators report what happened in each zone; analyze errors and their root causes rather than listing symptoms; then issue the overall verdict. In U.S. practice aligned with the exercise evaluation approach, separate observed strengths and areas for improvement from an improvement plan that names an owner and a deadline for each corrective action.
Keep a drill log with the date, scenario, timestamps, verdict, and open items. Before judging progress, verify that corrective actions from the previous drill were actually closed. A drill that simply repeats last quarter's unresolved flaw is not improvement — it is evidence that the review loop is broken.
- Debrief immediately, with all participants present.
- Analyze causes of errors, not just the errors themselves.
- Log results and issue an improvement plan with owners and deadlines.
- Check that prior corrective actions are closed before scoring progress.
Common scoring mistakes and how to avoid them
Several recurring mistakes undermine drill scores. First, judging from memory without written timestamps and a completed criteria sheet. Second, comparing your time against an external benchmark unrelated to your building and occupant profile. Third, grading speed alone while ignoring decision quality. Fourth, running a compliance-focused debrief that never examines why errors happened or what to change.
The fixes are inexpensive: give every evaluator the same scorecard and a two-minute briefing before the drill, assign a dedicated timekeeper, and reconcile the observers' notes before anyone issues a verdict. Remember that the purpose of scoring is not a number for the report but an accurate signal of where the plan works and where it needs to change.
Put it into practice
Emergency Drill Scorecard: Speed × Decision Quality
A reusable weighted scorecard for lead evaluators and observers. Rate each criterion 0 (not done), 1 (partly done), or 2 (done), multiply by the block weight, and total. A listed critical error voids the drill regardless of score.
- Speed — alarm-to-first-movement: 0 if more than twice target, 1 if up to twice, 2 if within target.
- Speed — time for the last person to reach the assembly point versus target.
- Speed — time to complete the headcount at the assembly point.
- Decisions — alarm clearly understood in all zones and action began per protocol.
- Decisions — alternate route used correctly when the scenario blocked the primary exit.
- Decisions — restrooms and utility rooms swept, doors closed but not locked.
- Decisions — assistance provided to people needing help or unfamiliar with the plan.
- Coordination — headcount reconciled against roster and visitor/contractor log.
- Coordination — discrepancies reported to the leader and all-clear issued correctly.
- Documentation — timestamps and scorecard logged, debrief held, prior corrective actions closed.
Questions people ask
What timestamps should I capture during an evacuation drill?
Capture at least three objective timestamps: time from the alarm to the first movement of evacuees, the time the last person reaches the assembly point, and the time the headcount is complete. Record them on the spot because recall degrades within minutes. Comparing these against a target you set before the drill yields the speed component of the score.
Does OSHA require a specific number of evacuation drills per year?
OSHA does not specify a fixed annual count of evacuation drills. What the standard does require is that your emergency action plan include evacuation procedures and that every employee understands their responsibilities; conducting drills is a practical way to satisfy and demonstrate that obligation. Many insurers and facility programs treat annual full evacuation plus periodic alarm tests as a reasonable baseline. Check your own insurance and industry requirements for any stricter expectation.
How should I account for people who are not employees during a drill?
Include them explicitly. Many plans count only staff, which produces a false headcount. Accountability procedures should cover visitors, contractors, and vendors, typically through a visitor log and a designated person responsible for those groups. Account for the people who may be in restrooms, utility rooms, or outside the building, and train designated sweepers to check those areas before leaving. The assembly area must be large enough for everyone and positioned so it does not interfere with arriving emergency responders.
Who should evaluate the drill, and why not just the leader?
Do not let a single person who is also running the drill grade it alone. Assign evaluators or observers to different zones or floors so someone watches each part of the movement. In practice, evaluators use the same scorecard, note timestamps and behaviours, and reconcile their notes before a verdict is issued. This reduces blind spots and makes ratings more objective.
What is a critical error that should fail a drill regardless of time?
A critical error is a decision that, in a real event, would escalate harm. Common examples are an unresolved headcount discrepancy where someone could still be inside, routing evacuees into a blocked or hazardous area, and ignoring the alarm entirely. Any critical error means the drill is not counted as successful, and you should analyze the cause and schedule a repeat before moving on.
How do I set a realistic target evacuation time for my building?
There is no universal figure, so derive the target from your own site. Base it on the design assumptions in your evacuation plan, the footprint and number of floors, the occupant load, and the mobility of the people present, then refine it with results from earlier drills. State an acceptable tolerance in advance. Comparing each new drill against this site-specific baseline is far more meaningful than comparing against an external round number.
What should happen after the drill to close the evaluation loop?
Debrief promptly with everyone who participated. Restate objectives, have evaluators report what happened in each zone, analyze the causes of errors, and issue the overall verdict. Log the date, scenario, timestamps, and verdict, then write an improvement plan that assigns an owner and a deadline to each corrective action. Before the next drill, verify those actions were actually closed so you are measuring real progress.
Sources and further reading
Sources were checked when this page was generated. Confirm changing dates, rules and prices with the original publisher.
- Drill Planning Checklist (IS-360)FEMA (Federal Emergency Management Agency)
- Fire Evacuation and Headcount FAQsJ. J. Keller & Associates
- Методические рекомендации МЧС России (30.06.2014), раздел 5.3 «Анализ (разбор) результатов тренировки»МЧС России
- Методические рекомендации «Организация тренировок по эвакуации персонала предприятий и учреждений при пожаре» (утв. МЧС России 04.09.2007 № 1-4-60-10-19)МЧС России
- VI. Разбор тренировок (методические рекомендации по противоаварийным тренировкам)КонсультантПлюс