Three agencies in one room, and three separate notebooks. The police inspector writes in a pocket book. The council duty officer types into a Word file on a laptop that gets shut down at the end of the shift. The fire service logs on its own system, which the council will never see. Everybody is recording. Nobody is keeping a record.
A multi-agency decision log is one chronological record of what was decided, who decided it, what information they had, and why the other options were rejected. Three parallel logs are not that record.
A multi-agency decision log is one chronological record of decisions, their rationale and the information available at the time, kept by the body coordinating the response. Separate agency logs record activity. A joint log records judgement, which is what a public inquiry, a coroner or a scrutiny committee asks about.
What Is A Multi-Agency Decision Log?
It means writing down each significant decision as somebody makes it, with the time, the named decision maker, the information in front of them, the options they weighed and the reason for the choice. An activity log records what happened. A decision log records why it happened.
The difference gets lost because both are called logs and both are kept in the same room. An activity log is a sequence of events: a call received, a cordon placed, a rest centre opened. Useful, and no evidence of judgement at all. A decision log holds the moments where somebody chose between courses of action on an incomplete picture. Those moments are what anybody reviewing the response will want to understand, and they are the least likely to have been written down. A decision log is the discipline of catching those moments while they are still happening.
Why Does A Multi-Agency Incident Record Matter Years Later?
Memory fades and the file does not. A multi-agency incident record becomes the main evidence of how judgement was used under pressure, often several years later. The people who made the decisions will be asked to explain them from the document, not from what they remember.
The document can also be compelled. Under section 21 of the Inquiries Act 2005, the chairman of a public inquiry may require a person by notice to produce any documents in his custody or under his control that relate to a matter in question at the inquiry. A decision log is one of those documents. So is the notebook, the group chat, the shared drive and the version of the log somebody tidied up afterwards. The question is never whether the record gets read. It is whether the record says anything useful when it does.
Who Owns The Decision Log When Several Agencies Respond?
The coordinating body owns the joint log, and each responding agency keeps its own alongside it. Most national frameworks set it up that way. Almost none of them say how the joint log actually gets written. The split itself is deliberate: a joint record for joint decisions, agency records for everything each agency does alone.
In Ireland, the Framework for Major Emergency Management, developed in 2005 and adopted by government decision in 2006, designates one of the three Principal Response Agencies as Lead Agency and gives it the responsibility and mandate for the coordination function. It names information management systems among the arrangements that have to be in place for coordination to work.
In the United Kingdom, the joint decision log is the business of whoever chairs the response. The Joint Emergency Services Interoperability Principles programme publishes a Joint Decision Log template, last updated in March 2022, to sit alongside its Joint Doctrine, Version 3.1 of April 2024. The Joint Decision Model in that doctrine runs through five stages, from gathering information and intelligence to taking action and reviewing what happened, around one overarching aim: to save lives and reduce harm.
The United States takes a different route. The ICS 214 Activity Log is a standard National Incident Management System form, currently at version 3.1, and the forms booklet describes it as an optional two-sided form completed by all sections and units. Every section keeps its own log and the Documentation function gathers them up.
The structures differ. The gap is the same in all three. Each one says who holds the record and none of them writes it, so that job falls to a person in a room with several other things to do.
What Must A Multi-Agency Incident Record Contain To Stand Up?
Four fields carry the weight: the time, the named decision maker, the information relied on, and the reason for the decision. Drop the fourth and what is left is a chronology, which proves sequence and nothing else. Rationale is the only field that cannot be reconstructed honestly afterwards.
The table sets out what each kind of record establishes and what it leaves open.
| Kind of record | What it establishes | What it leaves open |
|---|---|---|
| Individual agency activity log | What that agency did, and in what order | Why it did it, and what the other agencies were doing |
| Joint decision log with rationale | The decision, the named decision maker, the options weighed and the reason | Whether the decision turned out to be the right one |
| Communications and acknowledgement record | Who was told, and who confirmed it | What was decided on the strength of it |
| Timeline reconstructed after the event | A plausible account of the sequence | Almost everything, because it records what people remember |
Councils rarely lack the first and third rows. The second is the one that goes missing, because nobody has been given the job and the people qualified to dictate it are the people running the incident. Crisis Management Software can timestamp and store a decision entry. It cannot think of the sentence that goes inside it, which is why a decision log is a staffing question before it is a software question.
How Does Decision Logging For Emergency Planning Officers Work In Practice?
It works when logging is somebody's only job. Decision logging for emergency planning officers fails for a predictable reason: the person best placed to record the reasoning is the person making the decisions, and they are fully occupied. Split the two roles and the record improves straight away.
- Name a loggist for every activation, with a named relief for anything longer than a few hours, and put both in the plan instead of deciding on the night
- Log decisions, not activity. “Evacuate the ground floor and open the leisure centre” is a decision. “Spoke to the fire service” is not
- Record what was known at the time, including what was not known, and name the source. A decision that looks wrong later often looks reasonable against the information available
- Record the options rejected and the reason why. This single entry answers the question an inquiry actually asks
- Timestamp at the point of decision, not at the point of writing up, and mark any entry added later as retrospective
- Hold the joint log where partner agencies see the same entries, so three accounts do not drift apart in the first hour
How Does Crises Control Help With A Multi-Agency Decision Log?
Crises Control handles the part of a multi-agency decision log that a platform can genuinely own: the structure, the sequence and the custody of the record. One workspace, one timeline, one set of entries that every partner agency can read.
Public Sector Incident Management Software brings departments, responders and partner agencies into one shared workspace, and its Operational Audit Trail maintains an incident timeline for reporting, compliance and post-incident reviews. Every alert, task and decision is recorded automatically as a time-stamped audit trail, so the joint log sits in one place instead of three.
Incident Manager, the coordination module described on the incident management software page, maintains a chronological record of communications, updates and decisions throughout every incident. Every communication, acknowledgement, update and decision is timestamped and retained as a complete incident record, and that record is exportable to support governance, compliance reporting and post-incident review. Digital attachments let the plans, procedures, maps and contact lists that informed a decision sit with the decision itself. Incident Reporting Software turns the same activity into an Executive Summary Report, a Detailed Incident Report or an Audit Trail Report, exportable as PDF or CSV.
What the platform does not do is supply the reasoning. There is no feature called decision logging, and no software can write the sentence explaining why the leisure centre was chosen over the school hall. What Crises Control does is make that sentence easy to capture, impossible to backdate and findable two years later. Somebody still has to type it. The same split runs through the alerting side of the response, which we looked at in proving who your council actually warned. The platform holds the evidence. The authority holds the judgement.
Decision Log Checklist For Multi-Agency Response
Work through this before the next activation, not during it. It is the short version of keeping a decision log that survives being read by somebody who was not there.
- Name the loggist role in the plan, with a relief, and train more people in it than you expect to need.
- Agree with partner agencies which log is the joint record and where it lives, before an incident makes the question urgent.
- Write the four fields into your template: time, named decision maker, information relied on, reason for the decision.
- Record rejected options as routine, not only when a decision feels contentious.
- Mark later entries as retrospective, and never overwrite an earlier entry.
- Keep the information that informed each decision attached to it, including the version of the plan in force that night.
- Test the export. Read last year's record back and ask whether it explains a single decision to somebody who was not in the room.
What The Inquiry Room Actually Tests
An inquiry rarely argues about whether a decision was correct. It asks what the decision maker knew, what they considered, and why they chose as they did. Those three questions are answerable from a decision log and close to unanswerable from anything else. So the quality of the log often decides how a review goes, more than the quality of the response does.
That is an uncomfortable thought, and it cuts both ways. A well-kept log protects officers who made reasonable calls on poor information, which is most calls in the first hour of a major emergency. The absence of a log protects nobody. It leaves the reconstruction to other people, years later, who already know how it turned out.
Treat the decision log as part of the response, then, rather than part of the paperwork. To see what a joint record would look like against your own multi-agency arrangements, book a demo and bring the debrief from your last major incident with you.
Frequently Asked Questions
What is a decision log in emergency management?
A decision log is a chronological record of the significant decisions taken during an incident, each entry naming the time, the decision maker, the information relied on and the reasoning. It is different from an activity log, which records events rather than judgement. In the United Kingdom the Joint Emergency Services Interoperability Principles programme publishes a Joint Decision Log template for multi-agency use.
What is the difference between a decision log and an activity log?
An activity log records what happened and in what order. A decision log records why somebody chose one course of action over another. Only the decision log answers the questions a public inquiry, a coroner or a scrutiny committee asks. Both are worth keeping. Most response teams keep the first well and the second badly.
Who is responsible for the decision log in a multi-agency incident?
The coordinating agency holds the joint log and each responding agency keeps its own. Ireland's Framework for Major Emergency Management assigns the coordination function to a designated Lead Agency, while the United States Incident Command System has every section complete an ICS 214 Activity Log. In practice the record only gets written if a named loggist is assigned to do nothing else.
Can a decision log be written up after the incident?
It can, and it is worth much less. A log reconstructed afterwards is evidence of what somebody remembers rather than evidence of judgement at the time, and reviewers treat it that way. Any entry added after the event should be marked as retrospective rather than presented as contemporaneous. Retrofitting a log is also where well-meant record keeping turns into a credibility problem.
What should each decision log entry contain?
Four fields: the time of the decision, the named person who made it, the information available to them and the reason they chose as they did. Add the options rejected and the entry becomes genuinely defensible. Incident reporting software that timestamps and retains entries automatically removes the step where somebody has to remember, though the reasoning still has to be written by a person.
This article was drafted with AI assistance and reviewed by the Crises Control team. Featured image: AI-generated.
