21 October 2026
An organisation can invest heavily in reliable technology and still struggle when a critical service fails. The difficult moment arrives when the dashboard turns red and people must decide what to do next.
Who takes charge? Which services must continue? When should a specialist supplier be called? What should customers be told while the cause remains uncertain?
These decisions connect technical recovery to organisational resilience. Secure systems, tested backups and skilled engineers are essential. Their value depends on an organisation being able to use them effectively under pressure.
A technical failure with organisational consequences
The disruption to UK air traffic services on 28 August 2023 provides a useful case study. A flight-plan processing failure at NATS affected airlines, airports and passengers. The Civil Aviation Authority later estimated that more than 700,000 passengers were affected. Its independent review was published in November 2024.[1]
The review explains that the same flight-plan data triggered failures in both primary and secondary processing systems. Manual processing was available, but at substantially reduced capacity. It also identified difficulties involving specialist support, escalation, documentation and overall incident accountability.[2]
Those findings matter beyond aviation. The existence of a fallback arrangement does not, by itself, demonstrate that an institution can sustain the service people need.
The review’s recommendations extended to contingency arrangements, engineering resources, earlier communication and rehearsing major incidents across the aviation sector.[1] This is a historical case study; those findings should not be read as a description of NATS’s current arrangements.
Recovery begins before the outage
For a bank, a technically successful recovery might mean restoring a payments application. For its customers, recovery means knowing whether their payments completed, whether they should retry them and when they can safely rely on the service again.
For a school, restoring a learning platform may leave teachers managing missed assessments. For a government agency, bringing a portal online may still leave a backlog of applications and people unable to access essential services.
The practical implication is to define recovery around the service and the people who depend on it. Technical restoration is a milestone. Reconciliation, backlog management and clear communication may remain unfinished.
This wider perspective also appears in banking guidance. The Basel Committee’s 2021 Principles for Operational Resilience focus on maintaining critical operations through disruption. They connect governance, continuity testing, dependencies, incident management and technology resilience within one approach.[3]
Cybernexa’s interpretation is that digital resilience becomes credible when these elements work together in practice.
Human capability needs organisational support
Describing resilience as a human capability should never become a way of blaming employees for failures they were poorly equipped to manage.
A service-desk analyst cannot escalate effectively without clear thresholds and reachable contacts. A branch manager cannot offer a safe alternative without knowing which activities are permitted. A communications team cannot provide useful updates without access to reliable operational information.
Training should therefore reflect the decisions each role must make. Employees need opportunities to practise recognising disruption, reporting uncertainty, protecting information and using approved alternatives. Managers need to practise prioritisation, delegation and decisions made with incomplete information.
Leadership must provide the conditions that make those skills useful: clear authority, adequate resources, accessible procedures and permission to raise concerns early. Training and organisational design should reinforce each other.
Applied Cybernexa Scenario™: The system returns, but the service has not recovered
The following is a fictional learning scenario.
A regional bank’s payment service becomes unavailable during a busy morning. Its technology provider begins recovery work while branches receive increasing numbers of customer enquiries.
After two hours, the application is restored. However, some transactions have uncertain status. Branch employees do not know whether to advise customers to retry payments, and the contact centre has received no agreed guidance.
Management now faces decisions that a system restart cannot resolve. Which transactions need reconciliation? Who can approve temporary restrictions? What can customers safely do? Who will decide that normal service has resumed?
An effective response would bring technology, operations, risk and customer communication together. It would establish a shared view of affected transactions, assign decision authority and give frontline teams consistent instructions. Recovery would include checking outstanding work and correcting customer harm where identified.
The exercise should assess how decisions are reached, whether information reaches the people who need it and whether service restoration is verified. A successful technical recovery alone would leave important questions unanswered.
Evidence leaders should ask to see
Cybernexa proposes five practical questions for leadership teams reviewing their readiness:
- Which services must continue?** Identify the people affected, the consequences of interruption and the minimum service that can realistically be maintained.
- Who can make urgent decisions?** Name the incident lead, deputies and approval authorities, including arrangements outside normal working hours.
- Have alternatives been tested under realistic conditions?** Include reduced staffing, unavailable suppliers and disruption to the communication tools normally used to coordinate a response.
- Can frontline teams give consistent guidance?** Test how information moves from technical specialists to branches, classrooms, service counters and customers.
- What proves recovery is complete?** Check transaction integrity, outstanding work and service quality, then assign owners and deadlines to lessons identified.
The answers should be supported by exercise records, observed performance and completed improvements. A policy’s approval date says little about whether people can apply it during an unfamiliar incident.
Trust is tested during disruption
Customers and citizens may accept that systems sometimes fail. What they experience during the interruption—and how the organisation puts things right—will influence whether they continue to trust it.
Digital resilience requires sustained investment in technology. It also requires people who can interpret events, exercise judgement and coordinate action, supported by an organisation that has prepared them to do so.
The question for leaders is practical: when a critical system fails, can the institution continue to make sound decisions and deliver the services people depend on?
That capability must be built and exercised before it is needed.
Sources & Further Reading
- UK Civil Aviation Authority, Aviation regulator publishes Independent Review into August 2023 NATS Flight Planning System Failure, 14 November 2024.
- Independent Review Panel, Independent Review of NATS (En Route) Plc’s Flight Planning System Failure on 28 August 2023 — Final Report, CAP2993, published by the CAA in November 2024. Relevant sections: paragraphs 2.9, 2.14 and 2.28; recommendations 1–8.
- Basel Committee on Banking Supervision, Principles for Operational Resilience, March 2021, particularly sections IV–V.