Friday the 13th makes a neat excuse to check the fire exits. The more useful question is whether anyone has checked that last night’s backup can actually be restored.
Disasters make memorable headlines. But many interruptions to UK SMEs begin with less cinematic problems: a finance server running out of disk space, a certificate expiring, a broadband line failing without a fallback, or a month-end process that only one person knows how to complete.
Superstition is not the risk. A risk register shaped by whatever feels dramatic, rather than what repeatedly happens, is.
The useful test is not “What would look alarming in a headline?” It is “What has happened before, and what happens when nobody is watching?”
Repetition is evidence
A routine failure can be easy to dismiss because it is familiar. The printer stops talking to the network again. A user cannot access a system after a password change. A workaround gets added to the month-end checklist, then quietly becomes part of the process.
But recurrence is not reassurance. It is evidence that the underlying cause remains.
This is where a helpdesk ticket history is more than a queue of requests. Reviewed properly, it can show which faults return, which sites or systems are involved, how much staff time they consume and whether fixes are resolving the cause or merely restoring service until next time.
Gut feel and length of service are not a risk assessment. A dated record of incidents, costs, causes and actions is much closer.
Disaster Recovery starts with a small test
A backup job that reports success tells you that a process ran. It does not, by itself, prove that the right information was captured or that anyone can restore it into a usable service.
Disaster Recovery becomes practical when the business checks its assumptions. Start small: restore one representative file, confirm it opens, check that the right people can access it and record how long the process took. Then walk through the runbook step by step. If a crucial instruction is unclear, or depends on a person who is unavailable, the test has found something useful.
Agree recovery objectives in business language. How long can the finance team be unable to invoice? How much recent data could the business afford to recreate? These answers help set a Recovery Time Objective and a Recovery Point Objective that reflect operational needs, rather than convenient technical settings.
A plan is a prediction until somebody has tired it out with a controlled test. The National Cyber Security Centre’s small business guidance is a useful starting point for preparing to respond and recover.

Cloud is a risk decision, not a fashion choice
An ageing on-premise server can put several business dependencies in one unremarkable place: a disk, a fan or a certificate in a cupboard at a site nobody visits regularly. Moving suitable services to managed Cloud Solutions can transfer some of the routine infrastructure work, such as patching and out-of-hours monitoring, to a provider whose role includes doing it.
That is not a promise that cloud services cannot fail. Nor does it remove the customer’s responsibility to understand access, data retention, recovery arrangements and supplier dependencies. The point is to assess whether the change reduces a fragile single point of failure and makes responsibility clearer.
Cloud should be an informed risk decision, not a technology fashion statement. An independent review can compare the business’s existing estate, recovery requirements and costs before a move is proposed.
Month-end should not depend on one workaround
An ageing finance system may appear reliable because staff have learnt how to work around its limitations. The risk can be hidden in accumulated customisation, undocumented integrations and a month-end process that only one employee understands.
A structured Dynamics 365 Jump Start can assess whether Microsoft Dynamics 365 Business Central is a suitable, supported and documented core for finance and ERP. That work should begin with processes, data, reporting, controls and continuity requirements, not a software demonstration.
The objective is not to move a workaround into a newer system. It is to understand which processes matter, document them, simplify where sensible and plan a controlled route to a system the business can support. As consultants, we advise on the fit and the implementation approach; we do not act as software resellers.

A dependable legacy platform can still carry an undependable risk
A legacy IBM i (AS/400) estate is a particularly clear example of how a reliable machine can sit inside a fragile operating model. The platform may run consistently, while the knowledge around it lives with one administrator, the support agreement has not been checked for years, or recovery steps exist only in someone’s memory.
The risk is not simply that the system is old. It is that the business may have no documented answer to who can support it, what it depends on, how it is recovered and which processes would stop if it became unavailable.
Specialist IBM i Management can turn that persistent dependency into a managed one: with documented procedures, clear support responsibilities, appropriate cover and a practical recovery plan. The aim is to protect the business process and make its support requirements visible, not to replace a system just because it has been around for a while.
Look for the quiet leaks in technology spend
Some technology risks never produce an incident because they never get examined. A licence renews for software nobody uses. Two mobile contracts cover the same need. Connectivity remains in place at a closed site. An after-hours support agreement turns out not to cover the hours the business assumed it did.
A quarterly, evidence-based Technology Expense Management review can match contracts and invoices to actual usage, owners and business requirements. It should check renewal dates, duplicate services, support terms and connectivity across sites.
This is not cost-cutting for its own sake. The aim is to make expenditure defensible, identify avoidable waste and direct resources towards the remediation that reduces real operational risk.
Execution services should support the strategy
Once priorities are clear, Managed IT Services, Helpdesk Support, Network Solutions and Voice and Connectivity can provide the operational coverage to put them into practice.
Network redundancy and connectivity failover are not glamorous. They can, however, remove a very ordinary cause of lost trading time: one failed line taking a site offline. Support arrangements also need to account for evenings and weekends, when a problem may be just as disruptive and fewer people are available to respond.
The Fractal IT Director approach puts strategic consultation first. David Evestaff’s practitioner background, including Google-certified IT project management, multinational onboarding and IBM i system administration, informs that advice. A trusted network of specialist channel partners can provide specific technical execution and scale where needed. The technology support is the bolt-on; the business strategy is the point.

Boring evidence is still evidence
ISO 27001 and Cyber Essentials are consultancy domains, not products to sell. Both reward clear evidence of how an organisation manages security, although they address different aspects of it. Cyber Essentials provides a baseline around core technical protections; it should not be mistaken for proof that a business can recover its services. NCSC guidance explains the scheme’s scope.
For an SME, useful evidence may include patching records, asset registers, backup restoration tests, incident logs, supplier reviews and documented continuity arrangements. These records help turn a stated intention into something that can be reviewed, challenged and improved. An ISO 27001 consultancy review can help an organisation understand how information security management applies to its context and objectives.
Three questions for this quarter’s log
Replace the Friday the 13th instinct with three questions about each undramatic failure recorded this quarter:
- What happened?
- What did it cost in lost hours?
- What would have prevented it?
The answers are often boring, repeatable and relatively cheap: a tested restore, a documented step, a renewed certificate, a second connection or a clearer support agreement. That is precisely why they belong on the risk register.
SEO tags: Cloud Solutions, Disaster Recovery, Dynamics 365 Jump Start, Helpdesk Support, IBM i Management, Managed IT Services, Network Solutions, Technology Expense Management, Voice and Connectivity, UK SME IT consultancy, business continuity, ISO 27001 consultancy, Cyber Essentials guidance
Join The Discussion