The emergency plan was complete. The first handover was not.
In a fictional regional library network, every branch has a continuity folder. It identifies an alternative location, lists staff responsibilities and describes how residents should receive information if their usual building becomes unavailable. At a rehearsal, the branch coordinator sends the required message. The receiving office does not know who is supposed to acknowledge it.
Nothing dramatic has happened. No real building has closed. No resident has been redirected. Yet the exercise has discovered something important: a written transfer of responsibility is not the same as a transfer that people can actually perform.
Emergency preparedness, readiness testing and continuity exercises help civilisation distinguish resources it possesses from capabilities it can demonstrate. A rehearsal should not exist merely to produce a reassuring report. It should expose the assumptions between a plan and a usable public outcome while the consequences of discovery remain controlled.
This casebook follows a fictional exercise from design through observation, evaluation, correction and retest. All participants, timings and success criteria are teaching constructions. The proposed activities use synthetic records and simulated communications. They do not require closing real services, triggering real alarms, using live personal data or interfering with operational systems.
A 50-second readiness test
Choose a bounded capability and describe the evidence that would demonstrate it. Select an exercise type suited to the claim. Make important assumptions visible. Let participants attempt the task without silently supplying missing steps. Record what happened, including help from facilitators. Separate demonstrated performance, observed failure and untested capability. Give each correction an owner and a retest. Accept only the readiness claim the evidence supports.
“The exercise finished” is a statement about the exercise. “The service is ready” is a stronger statement about the world. The second requires a defined scope, appropriate evidence and a clear account of what was not tested.
Choose a reading route
Part I — Turn a preparedness claim into an observable question
Part II — Design an exercise that cannot injure the real service
Part III — Watch the fictional rehearsal unfold
Part IV — Judge readiness without turning uncertainty into a score
Part V — Repair, retest and transfer the capability
For the wider management of launching and handing over projects, use Project Readiness Management. The question here is narrower: what can a civilisation learn by rehearsing a disrupted public-service journey before a real event forces it to improvise?
Part I — Turn a preparedness claim into an observable question
A folder is an asset; a working response is a capability
The fictional library network has bought equipment, written instructions and assigned names. These are useful preparations. They do not establish that the equipment is reachable, that staff understand the instructions or that a named person will be available when needed.
That distinction can be tested without dismissing the preparation. Ask what outcome the assets are meant to enable. For this exercise, the network chooses a modest function: when one branch is simulated as unavailable, authorised staff can establish an alternative information service and provide an accurate, accessible message through two designated channels.
The claim is deliberately smaller than “the region is ready for emergencies.” It does not cover structural safety, medical response, evacuation or every possible loss of infrastructure. Those require other expertise and exercises. Narrowing the claim makes the test more meaningful because participants know what evidence they are trying to produce.
Civilisation-wide preparedness grows from such connected capabilities. A broad promise becomes credible only when the component jobs and their handovers can be demonstrated, rather than assumed to work because each institution has a document.
Choose verbs that can leave evidence
Words such as understand, coordinate and communicate can hide very different expectations. In a readiness exercise, replace them where possible with observable actions. Locate the current contact route. Send the correct synthetic case identifier. Receive acknowledgement from the designated role. Produce a consistent message for each channel. Record who owns the next step.
This does not reduce good judgement to a mechanical checklist. It gives observers a basis for describing what happened. An evaluator can see whether a recipient accepted responsibility; they cannot directly observe whether everyone felt coordinated.
A strong objective also specifies the conditions. Is the usual coordinator present? Are normal digital tools available? Is the information complete? Must the participant work independently, or is the purpose to practise with coaching? These distinctions change what the result demonstrates.
The same action can therefore support different conclusions. Finding a number after a facilitator points to the correct page is useful training. Finding it independently in an unfamiliar but realistic case is stronger evidence of readiness. Neither should be reported as the other.
Select the exercise type to match the claim
WHO distinguishes discussion-based exercises, such as tabletop exercises, from operations-based activities including drills, functional exercises and field or full-scale exercises. Its description emphasises testing assumptions, roles, resources and coordination in simulated conditions. This is a health-emergency framework; the library case uses the general distinction without adopting clinical procedures. [1]
A tabletop conversation can reveal whether managers agree about responsibility. It cannot prove that a real communications path delivers messages under load. A carefully isolated drill can demonstrate a specific action, but may not reveal how several institutions coordinate over a prolonged disruption.
The library therefore begins with a discussion to expose disagreements, then tests selected communication and handover tasks using clearly labelled exercise channels. It does not jump directly to a spectacular full-scale event. The chosen format follows the evidence needed.
This is a useful antidote to exercise theatre. A large rehearsal is not automatically better. Complexity is justified when it exposes a meaningful dependency that a simpler exercise cannot test.
Write down what the exercise is allowed to conclude
Before anyone begins, the team writes an acceptance statement: the exercise will examine the designated daytime handover, production of a consistent public-information message, and access through two specified channels using synthetic requests.
Alongside it, the team records exclusions. It will not establish overnight staffing readiness, physical suitability of the alternative venue, performance during a real network outage or compliance with every legal obligation. Those exclusions are not small print added after a disappointing result. They define the evidence boundary from the start.
This protects against a subtle reporting error. After a successful narrow exercise, a summary can expand the claim until it sounds as though the whole regional system has been certified. The underlying observation has not changed; only the language has grown.
A readiness statement should travel with its scope. The reader should be able to answer: ready to do what, under which conditions, demonstrated by whom, using which version of the arrangement, and with what remaining uncertainty?
Part II — Design an exercise that cannot injure the real service
Realism should come from decisions, not uncontrolled disruption
A realistic exercise does not require frightening the public or disabling a working service. The library case creates pressure through incomplete information, a missing acknowledgement and competing tasks. It does not manufacture physical danger.
Exercise communications are clearly labelled. Synthetic names and records are used. Staff know which channels are for rehearsal and which remain operational. The plan identifies who can pause the exercise if real work needs attention. A real emergency takes precedence over the simulation.
These boundaries do not make the learning trivial. Participants still need to locate current instructions, recognise the limits of their authority, produce a correct message and negotiate a handover. Those cognitive and organisational tasks are precisely what the exercise is designed to examine.
The principle is to preserve the decision problem while controlling unnecessary consequence. Disruptive testing of actual critical infrastructure, clinical services or security systems requires specialised authorisation and safeguards far beyond this educational casebook.
Keep the scenario separate from the evidence
The scenario states that one branch is unavailable. This is an assumed event, not a finding. The exercise then observes how participants respond to that assumption. If the alternative venue is declared usable by the scenario, the exercise has not tested whether the venue is actually usable.
Write scenario assumptions where evaluators can see them. Which resources are assumed available? Which travel times are assumed? Which messages are delivered by the facilitator rather than the real communication route? Each assumption removes a possible failure from the test.
That can be appropriate. An exercise focusing on responsibility does not need to test every technical subsystem simultaneously. The error is allowing the assumption to reappear in the final report as demonstrated capability.
In the library example, the alternative room’s availability remains an assumption requiring separate confirmation. The message handover is tested. Those two states should remain distinguishable even if the rehearsal feels smooth overall.
Separate participants, facilitators and evaluators
Participants perform the simulated response. Facilitators keep the exercise within scope and introduce information at the intended moments. Evaluators record actions against the objectives. One person can sometimes hold more than one role in a small learning exercise, but that overlap should be acknowledged because it affects the evidence.
FEMA’s HSEEP training material describes exercises as a way to assess capabilities and identify strengths and gaps. It connects objectives, observation, evaluation and corrective actions rather than treating the event itself as the final product. Its training framework also distinguishes exercise-planning and delivery roles. [2]
For our case, a facilitator who wrote the procedure does not quietly coach participants and then present an independent verdict that the procedure worked. Coaching can be valuable; it must be recorded. A later uncoached attempt can test whether the learning transferred.
The point is not to pretend that every community exercise has a large independent review team. It is to avoid overstating review independence and to make potential sources of favourable interpretation visible.
Use a few discriminating events rather than a flood of surprises
The library exercise needs only a small set of events. First, the primary branch is simulated as unavailable. Second, the usual recipient does not acknowledge the message. Third, a synthetic resident cannot use the digital channel. Fourth, an updated location detail arrives after the first draft notice has been prepared.
Each event tests a different assumption: continuity ownership, timeout handling, access, and version control. The scenario is not designed to overwhelm participants with every imaginable complication. Too many simultaneous events would make it harder to identify why a task failed.
For each event, the designer records the intended observation. When the acknowledgement is missing, does the sender recognise the gap and use the documented alternative? When the location changes, do both public-information channels receive the same correction?
A useful exercise creates a situation in which an important weakness would become visible. Surprise is a tool for that purpose, not a measure of exercise quality in itself.
Part III — Watch the fictional rehearsal unfold
The first message leaves; responsibility does not arrive
At the beginning, the branch coordinator sends a correctly labelled exercise message to the regional office. The record shows a successful send. No acknowledgement follows. Participants initially describe the handover as completed because they performed the action listed in their folder.
The evaluator asks no leading question. Instead, the record distinguishes message sent from responsibility accepted. The receiving office is unsure which duty role should answer. Both sides can explain their local actions, but the service still lacks an accepted next owner.
This is a productive failure. It reveals an interface defect before a resident depends on the arrangement. The correction should not merely tell the sender to be more careful. It should define the accepting role, the evidence of acceptance and the route used when acknowledgement does not arrive.
The case connects to coordination across system boundaries. Two adequate local procedures can leave an inadequate shared process between them. The exercise makes that gap observable.
Three route timings tell a different story from their average
Suppose the team repeats three synthetic handover attempts. Before the exercise, it chose a four-minute acknowledgement target for this educational scenario. The observed times are seven minutes, two minutes and five minutes.
The mean is fourteen divided by three, approximately 4.67 minutes. That average does not describe any one route. One attempt met the target; two did not. Reporting only a typical response would conceal which handovers remain unreliable.
| Attempt | Observed time | Target met? | What must be investigated? |
|---|---|---|---|
| First | 7 minutes | No | Receiver ownership was unclear |
| Second | 2 minutes | Yes | Whether the successful route is repeatable |
| Third | 5 minutes | No | The alternative contact route added delay |
These three observations are too few to estimate a dependable population-wide success rate. They are still sufficient to identify two witnessed misses against the stated exercise criterion. A small test can reveal a defect without proving how frequently that defect occurs in real operation.
The four-minute target is invented, not a public-service or emergency standard. In an actual setting, the acceptable time must be justified by the function, consequences and authorised requirements.
A digital message is correct; the accessible route is incomplete
The team produces a clear digital notice. It identifies the temporary service, its limits and the next update. A synthetic resident then asks for the same information through the assisted channel because they cannot use the website.
The assisted-channel staff have an older version of the location detail. No one has deliberately withheld information. The update route simply reached one channel before another. The service would therefore offer different instructions depending on how a resident contacted it.
The exercise record should preserve that distinction. It would be inaccurate to say public communication failed completely; the digital notice was correct. It would also be inaccurate to say communication passed because one channel worked. The capability being tested included consistency across both.
This kind of finding is especially valuable at civilisation scale. A technical system can be functioning while practical access remains unequal. Testing only the easiest user journey would leave the excluded route invisible.
Coaching changes the state of the evidence
After the first failure, the facilitator helps staff identify the accepting role and the correct update path. Participants then complete the tasks. The rehearsal has produced learning, which is a legitimate outcome.
It has not yet demonstrated that another shift can perform the revised procedure independently. The report should say what was achieved with assistance and what remains to be retested without it.
There is no reason to shame participants for needing help in a learning exercise. The problem would be reporting assisted success as proof that the original arrangement was already ready. That erases the defect the rehearsal was meant to uncover.
A good record therefore notes intervention by facilitators: what prompt was supplied, when it was supplied, and which subsequent action depended on it. The next exercise can deliberately remove that help and examine whether the corrected system now supports ordinary performance.
Part IV — Judge readiness without turning uncertainty into a score
Passed, failed and untested are different states
Suppose the exercise examines four required capabilities. Three are demonstrated under the specified conditions. The fourth cannot be tested because the external partner is not participating. It would be misleading to announce that the network is 75% ready.
The denominator counts capabilities, not comparable units of preparedness. One untested function might be essential to all the others. Nor should untested be relabelled failed. The exercise did not observe its failure; it lacked evidence about it.
A better statement is precise: three named capabilities were demonstrated in the exercise; the external-partner handover was not tested; readiness for the complete journey remains unconfirmed. That language preserves both achievement and uncertainty.
The distinction is non-compensatory. Excellent performance on a notice-writing task cannot establish that a missing access arrangement will work. Where a capability is required for the claimed service, evidence about a different capability cannot fill the gap.
One successful shift does not represent every shift
The daytime team may know the procedure well because it helped write it. An evening or weekend team may have different staff, authority, suppliers and contact routes. A rehearsal with the first group cannot automatically represent the second.
This is a sampling problem as well as a staffing problem. State who participated and why that group was selected. When the aim is training, a motivated group is reasonable. When the aim is assessing routine readiness, exclusively selecting the strongest participants may overstate ordinary performance.
The next step need not be to test every possible person in every possible event. Identify materially different operating conditions and decide which need separate evidence. Out-of-hours ownership, reduced staffing and unavailable communications may be more consequential than minor variations in the scenario wording.
A civilisation becomes better prepared by knowing where its readiness evidence transfers and where it does not. Universal language supported by a narrow sample creates confidence without corresponding capability.
Technical success and institutional permission must remain separate
A participant may discover a technically effective workaround. That does not automatically mean they are authorised to use it in a real event. The workaround might require permissions, data access, expenditure or professional judgement outside their role.
The exercise should preserve the finding without silently changing authority. Record the proposed method, why it helped and which approvals or safeguards would be needed before actual use. An exercise can discover an option; it cannot manufacture legal or professional permission by demonstrating that the option is possible.
This distinction also protects staff. They should not be congratulated in rehearsal for a workaround they could later be punished for using operationally. The institution needs to resolve the mismatch between expected outcome and permitted action.
The relevant question is whether the ordinary role has enough authority to deliver the intended function safely. Where it does not, the repair belongs partly to governance, not merely to training.
A readiness judgment should be a bounded argument
After the fictional exercise, the evaluator assembles a short argument. The service claim is stated. Each necessary capability is matched to observations. Exceptions and facilitator interventions are recorded. Untested conditions are listed. The conclusion identifies what can currently be relied upon and what should remain conditional.
For example: the revised daytime message workflow was completed with coaching, but the alternative contact route exceeded the exercise target and the external-partner handover was not tested. The network should not claim complete readiness for the whole journey until those gaps are addressed and retested.
This is not a verdict that the institution is generally incompetent. It is a specific statement about a specific configuration of people, procedures and tools. The value lies in how directly it points to repair.
Public confidence is better served by such bounded honesty than by a ceremony announcing that all systems are ready. The institution can explain progress without implying that a finite exercise has proved success under every future condition.
Part V — Repair, retest and transfer the capability
Correct the mechanism, not the wording of the finding
The first finding concerns missing acceptance of responsibility. Rewriting it as improve communication does not make it actionable. The correction needs to specify the accepting duty role, the acknowledgement format and the fallback route when no acknowledgement arrives.
The second finding concerns inconsistent information across channels. Its repair might establish one controlled message source and a check that both channels carry the same version. The mechanism is different, so it deserves a different corrective action.
Each action needs an owner with the authority and resources to complete it. A deadline without ownership is merely a date. Ownership without a completion test allows the action to be closed when somebody says it has been addressed.
This is the link to learning from failure without repeating it. The exercise creates value when its evidence changes the next version of the system.
Retest the changed capability under a meaningful variation
Repeating the exact exercise immediately after coaching can measure short-term recall. That is useful, but limited. To test transfer, the next attempt might use another synthetic case, a different receiving role or a later shift, while keeping the core capability and evaluation criteria stable.
The variation should be meaningful rather than arbitrary. Changing every condition at once makes it difficult to attribute improvement or failure. A focused retest asks whether the corrected handover works without the facilitator supplying the missing instruction.
Record the version being tested. If the procedure, contact route or staffing arrangement has changed, the previous result belongs to an earlier configuration. This is not bureaucratic fussiness; it prevents an old passing test from being used to support a materially different current system.
The desired result is not endless testing. It is enough relevant evidence to support the next justified commitment, with a sensible plan for checking again when conditions change.
Restore the organisation after the exercise
A rehearsal can itself leave work behind. Temporary accounts, synthetic records, draft messages, borrowed equipment and altered calendars need to be closed or returned. Staff also need time to catch up on routine work displaced by participation.
The exercise closeout should therefore distinguish learning outputs from operational residue. Keep the approved evidence and findings. Remove or clearly isolate synthetic material so it cannot be mistaken for a real case. Confirm that no exercise message remains in a public-facing queue.
This is another reason to keep the rehearsal bounded. An exercise that leaves ordinary services confused has created avoidable harm while trying to improve preparedness. The clean end state should be part of the design, not an afterthought.
Closing the event does not close every corrective action. Those actions move into ordinary management with explicit owners and review dates. Otherwise the exercise team disperses and the most important findings lose responsibility.
Readiness can decay after a successful test
People leave. Contact routes change. A supplier replaces its system. A venue becomes unavailable. The procedure that passed a rehearsal may no longer describe the current arrangement.
A useful readiness record therefore names the changes that should trigger reassessment. A new colour scheme in a notice may not matter. A different identity system, a changed duty roster or removal of the alternative channel might materially affect the capability.
Review should also be proportionate to the consequence of failure. The criticality of the protected function helps determine where stronger and more frequent evidence is warranted. No single exercise calendar fits every public service.
The aim is maintained readiness, not a certificate that never expires. A civilisation should know which preparedness claims remain supported and which have become uncertain because the world changed around them.
A compact exercise dossier
A practical dossier can contain seven short sections: the capability claim, exercise scope, assumptions, planned observations, actual evidence, corrective actions and the next review. Each section should answer a real question. The dossier need not grow into an unreadable manual.
For the fictional library network, the capability claim concerns continuity information after a branch closure. The scope covers daytime staff and two channels. The assumptions include the alternative room’s availability. The observations include message acceptance, consistency and access. The evidence distinguishes the three route timings and the coached completion. Corrective actions address ownership and versioning. The next review identifies an uncoached retest and an external-partner check.
A reader can now reconstruct why the judgment was made. They do not have to rely on the evaluator’s enthusiasm or a single percentage. The dossier connects a claim to observable performance.
That connection is the real output of readiness testing. It turns preparedness from a collection of possessions into an inspectable argument about what the institution can do.
A classroom version that stays safely inside the classroom
Students can test a fictional school-library reservation handover. One group receives synthetic requests, another confirms availability, and a third prepares messages. Use invented names and clearly marked exercise documents. Do not send messages to real parents or alter a live school system.
Agree beforehand on the outcome: a request reaches the next group with the correct identifier and an acknowledged owner. Run one coached practice, then a separate uncoached attempt. Introduce one change, such as an unavailable recipient, and observe whether the agreed fallback is used.
Ask students to write three different statements: what the plan said, what they actually observed, and what they can reasonably conclude. They should mark untested conditions rather than filling them with confidence.
The best report may contain an honest failure. A discovered gap, a specific correction and a successful retest teach more about preparedness than a flawless performance that depended on the teacher quietly solving every problem.
The purpose of rehearsal is to make the next response less imaginary
The fictional network began with a complete folder and an incomplete handover. Its exercise did not prove readiness for every emergency. It did something more useful: it exposed a specific gap, showed how channel information could diverge and identified the limits of the evidence.
The improved system has a clearer accepting role, a controlled message version, an explicit alternative route and a retest plan. Its preparedness claim is narrower than the original slogan but better supported.
Civilisations test readiness well when they allow reality to disagree with the plan in a controlled setting—and then require the correction to work without the exercise quietly helping it along.
Sources and evidence boundaries
[1] World Health Organization, Simulation exercises. Used for the distinction between discussion-based and operations-based exercises and their role in examining preparedness assumptions. Health-specific procedures are not transferred into this library case.
[2] FEMA National Disaster and Emergency Management University, Homeland Security Exercise and Evaluation Program training material. Used for the relationship among objectives, exercise roles, capability evaluation, findings and corrective actions. This article is not an accredited HSEEP course or an official exercise evaluation.
The library scenario, four-minute target, route timings, readiness states and dossier are original educational constructions. An exercise provides evidence only within its design and observed conditions. Actual emergency operations require the applicable professional, legal and organisational authority.
Continue through the Civilisation Atlas
Return to the Civilisation Atlas, study coordination between experts and generalists, or connect the exercise findings to priority-setting.
Next casebook: Test the order of the recovery work itself in How Civilisations Restore Essential Services in the Right Order, where dependencies and one scarce specialist produce three different public-service outcomes.