Skip to content

Building and sustaining the UK's national emergency coordination platform

Resilience Direct sits at the heart of the UK's critical national infrastructure. It's the platform that coordinates emergency response across every police force, fire brigade, ambulance service and local authority in the country. CDS has been its primary delivery partner for over twelve years. 

UK_Cabinet_Office_Briefing_Room_(COBR)_Logo

OVER

1M

documents in one national repository

OVER

1TB

of secure file storage

UP TO

130K

registered users across the UK

Approx

8K

organisations on the platform

Overview & summary

CDS has been the primary delivery partner for Resilience Direct since the contract was won in 2013 - a twelve-year continuous improvement partnership maintained through a structured cycle of identification, specification, delivery and review.

The work spans people, process, data and technology: organisational homepages that orientate users under pressure, automated account provisioning, a category-based security model enforced across an exceptionally varied user base, and a modernised GDS front end on AWS-hosted infrastructure.

Throughout, CDS has acted as the de facto technical authority and trusted partner, across a multi-supplier ecosystem.


At a glance

  • Built RD Collaborate - live since 2014, cloned from One HR, hosted on AWS
  • Consolidated a fragmented national estate into 1M+ documents in one repository
  • Introduced organisational homepages & a "My Organisation" view for instant orientation
  • Automated user account creation via Amazon Cognito - previously 100% manual
  • Secured the platform with MFA across all 130k users and a Cat 1–7 access model

 

ORGANISATION PROFILE

Resilience Direct

COMMISSIONED BY: Cabinet Office - now operating under COBRA

PURPOSE: National emergency coordination & resilience planning

REACH: Every police, fire, ambulance, local authority & government department in the UK

USERS: 130,000 registered

USER BASE: Cat 1 vetted professionals through Cat 7 community volunteers

ESTATE: 1M+ documents . 1TB+ storage . ~8,000 organisations

A national platform commissioned to centralise emergency coordination across the UK - serving two distinct but deeply connected functions.

The first is a national repository for emergency planning documentation: the plans, procedures and protocols that define how the UK responds to major incidents involving critical national infrastructure — power stations, chemical facilities, transport networks and nuclear sites — maintained by resilience planning forums in every local authority.

The second activates when an incident occurs. RD Collaborate enables real-time coordination across multiple agencies simultaneously — assigning tasks, tracking progress, and providing a shared operational picture, with integrated GIS mapping through Airbox so responders can visualise incidents geographically and coordinate resources on the ground.

This is a critical part of national infrastructure which means a failure would not just be a technology problem, it would be a public safety problem.

A man standing on a tree stump looking out at the landscape and clouds before him
A woman on a chair above the clouds

The challenge

 Before Resilience Direct existed, the UK had no national coordination mechanism for major emergencies. Hundreds of authorities and services maintained their own plans independently - different formats, different storage, no shared access. When an incident crossed jurisdictional lines, there was no overriding command authority and no shared operational picture. The challenge is best understood across four dimensions. 

People

Orientation at the moments that matter most 

Emergency response requires collaboration between police, fire, ambulance, environmental agencies, air traffic control and voluntary services. Each operates independently and engages with the platform at very different levels of frequency. Many users only interact with RD during or immediately before a high-pressure incident.

The challenge for these people was not productivity in a conventional sense. It was whether they could orientate, find what they needed, and act confidently at the moments that matter most.

As the platform grew in scope over many years, navigation became a genuine barrier. New users could arrive and feel lost, unable to find content relevant to their organisation without traversing the full breadth of a system spanning every government department and emergency service in the country. There was no dedicated organisational view. Speed and clarity under pressure were precisely what the system was failing to deliver.

Process

No shared operational picture 

Before RD, emergency coordination was fragmented and inconsistent. Agencies communicated through phone calls, informal conversations and bilateral exchanges e.g. police with fire, ambulance with police. There was no shared record of activity and no visibility of what other agencies were doing.

During a major incident such as a multi-vehicle motorway collision, each organisation would make assumptions about what others were handling. Resources were duplicated or misallocated, and critical tasks could fall between the gaps.

Data

No single source of truth and a security risk 

Before RD, emergency plans were held locally by individual authorities with no national visibility, no standardised format and no single source of truth. Over time, the platform's own data estate developed its own challenges: multiple disconnected applications with separate user datasets and no unified taxonomy. This drove manual administration, duplication and risk of error. Managing users across multiple systems required repeated effort from a small team.

The security of that data posed a direct challenge. During a MFA implementation project with a third party, CDS identified critical vulnerabilities - one which the team estimated could be compromised within thirty minutes. For a platform now operating directly under COBRA, a breach could mean a national security incident.

Technology

A 20-year lineage, many suppliers

RD Collaborate has been live since 2014, carrying a codebase with lineage tracing back over twenty years through One HR and the EPIMS/Horizon platform. The front end was restyled to Government Digital Service (GDS) standards.

The platform operates as a suite of tools rather than a single integrated application, managed across multiple suppliers who do not always operate to a consistent standard. Changes to one component must be coordinated with all others. The multi-supplier ecosystem creates integration risk at every point where the systems meet.

The objectives

Our solution

CDS has been the primary delivery partner for Resilience Direct since 2013. The solution is not a single project. It is a continuous improvement partnership, sustained through a structured cycle of identification, specification, delivery and review.

Just as the challenge spanned people, process, data and technology, so does the solution.

People

Organisational homepages & “My Organisation”

The primary people-facing improvement was the introduction of organisational homepages and a “My Organisation” feature - giving each of the hundreds of organisations on the platform its own dedicated space, accessible immediately upon login, with their plans, active responses and tasks clearly organised under their own identity.

For a user arriving during a live incident, this replaced a disorienting, search-heavy experience with immediate, role-relevant orientation. Alongside platform improvements, CDS has operated as a trusted advisory partner to the RD team, giving  guidance on security, technical direction and cross-supplier coordination well beyond the contracted scope.

Dedicated space per organisation, available the moment a user logs in.
Plans, active responses and tasks organised under the user's own identity.
Advisory partnership built on twelve years of institutional knowledge and earned trust.

 

Process

Automation and continuous improvement

A significant process improvement was the automation of user account creation — previously a 100% manual process managed by RPS, the frontline support team. Automation via AWS Cognito removed this administrative bottleneck entirely, freeing RPS staff for higher-value support work and reducing the risk of error in provisioning.

The delivery model itself is a continuous improvement cycle. Ideas emerge through regular conversations between CDS and the RD team or via the national user group backlog; CDS proposes solutions informally, then produces a formal written specification through multiple review iterations. Work is costed, developed, tested and released on a scheduled basis.

User account creation moved from 100% manual to automated via Amazon Cognito.
Frontline support (RPS) freed for higher-value work, with less provisioning error.
Structured cycle: identify → specify → cost → build → test → release.

Data

Shared taxonomy & security leadership

CDS began aligning the organisational taxonomy and user datasets across Collaborate and Airbox — establishing shared schemas and data standards for future integration across what had been entirely separate systems. The category-based security model (Cat 1–7) was implemented and enforced, ensuring every user sees only what their clearance level permits.

During the client's MFA project with a third party, CDS identified critical vulnerabilities in the infrastructure. We wrote the correct MFA integration code ourselves, then shared it with Airbox and Zeal so they could implement compatible solutions. This was outside CDS's contracted scope, but it was the right thing to do for a platform of this national significance.

  • Shared schemas and data standards aligned across Collaborate and Airbox.
  • Cat 1–7 security model enforced across the entire user base.
  • Correct MFA integration written by CDS and shared with other suppliers.

Technology

AWS migration & a GDS front end

The core technology stack is RD Collaborate - a CDS-built and maintained web application hosted on AWS, with GIS/mapping integration through Airbox, a learning management system via Zeal, and user identity management through Amazon Cognito. MFA has now been implemented across the entire platform ecosystem.

The front end was restyled to GDS/Gov.UK standards, improving accessibility and Cabinet Office branding alignment. The platform was migrated from legacy EPIMS hosting onto its own AWS-hosted infrastructure - a stable, scalable foundation from which to continue incremental improvement.

  • RD Collaborate on AWS, integrated with Airbox (GIS), Zeal (LMS) and Cognito (identity).
  • MFA implemented across the entire platform ecosystem.
  • Migrated off legacy EPIMS hosting; front end aligned to GDS/Gov.UK standards.

The outcomes

The Cabinet Office now has a single national view of emergency response activity and resilience planning across the UK , on a platform that is continuously improving and answerable to the highest level of government security scrutiny. 

Why CDS

 The Cabinet Office already knew CDS through EPIMS, a mandated government application used by every department in the UK. When the mandate arose after the Selby disaster, CDS had the platform, the security model, the track record and the relationships. What has kept CDS on the account for twelve years is a more revealing question.

Additional reading

Two optional sections for the technically curious and the delivery-minded. Skip them if you've got what you need!

Technical deep dive

~ 4 min read · architecture & migration

Architecture & stack


RD Collaborate is a CDS-built and maintained .NET web application, hosted on AWS within a Cabinet Office cloud tenancy. Its lineage traces back through One HR and the EPIMS/Horizon platform — over twenty years of continuous government application development. The platform integrates with Airbox (GIS mapping), Zeal (learning management) and Amazon Cognito (user identity), with first-line support provided by RPS.

The application was migrated from its original UKFast hosting environment to AWS, with DDoS protection and web application firewall capabilities provided via AWS-native tooling.

Performance improvements

Following user complaints about page load times and search speeds, CDS conducted a comprehensive performance audit — testing and benchmarking the platform before producing a prioritised set of recommendations. Phase 1 was released on 23 February 2026, with Phase 2 following in March 2026.

Audit log optimisation

A planned improvement to the audit and email log system moves historical records into an archive table, so standard searches return results from the most recent 6–12 months only. Analysis showed information older than a few weeks is rarely accessed in practice - improving general audit performance without meaningful impact on day-to-day usage.

Our ways of working

~ 3 min read · how the partnership runs

How work flows

CDS and the RD team operate a continuous improvement partnership, iterating the platform in response to user feedback, the national user group backlog, proactive consultancy from CDS, and evolving government requirements. Improvement ideas emerge through two main routes: direct conversation between CDS and the RD team, and items submitted through the national user group backlog.

CDS initiates a meaningful proportion of the work proactively, identifying opportunities early on, and bringing formal proposals to the table. Once an idea is agreed, CDS produces a formal written specification, broken down into development, test, database, front-end and PM tasks to produce a cost estimate. Work is then developed, tested and released on a scheduled basis.

Cross-supplier technical leadership

RD operates across a multi-supplier ecosystem: Collaborate (CDS), Airbox, Zeal, AWS and the infrastructure layer. CDS is not formally responsible for the other suppliers' applications, but has consistently taken on a coordinating and quality-assurance role across the ecosystem. The RD team consults CDS informally on decisions, on security questions, and on technical strategy.

CDS's position as the de facto technical authority across the RD supplier ecosystem is a function of twelve years of consistent delivery and earned trust.

Governance & account structure

The account is governed through a combination of formal delivery structures and a relationship model built on direct, honest communication. Annual support covers ongoing maintenance, incremental improvements, and a full-time expert dedicated to the account.

The way of working enables flexible budget allocation. This means small pieces of work can be scoped, agreed and delivered quickly without going through the full estimation and release process. That flexibility is an important part of how the relationship works in practice.

Running a critical service that can't afford downtime?

We'd be glad to talk about what good looks like, whether you're at the start of replacing a legacy platform, looking to modernise and innovate, navigating a regulatory deadline, or trying to bring a non-technical team along for the ride.

Tell us a little about you and we'll come back within one working day.