January gets the reputation. December does the rehearsal, with the audience already waiting.
For a business whose financial year closes on 31 December, the month brings two demanding processes into close proximity: the December payroll run and the year-end close. Bank holidays compress working days, so payroll preparation and approval often move forward. Meanwhile, journals, reconciliations and adjustments continue towards the close. The payroll file may need sign-off while ledger work is still under way.
This is not simply a busy calendar. The processes can depend on the same finance system, overnight processing, backups and small group of people. There may be only one week in which to complete both, with little room to repeat a run or recover from a mistake. The practical question for a finance director in mid-November is not just, “Are we ready for year-end?” It is: “Can this IT estate run two critical processes in the same fortnight without either one needing to be rerun?”
That is a business continuity question, and it deserves a business answer. Map the sequence: payroll inputs, checks, approvals, file production, payment and ledger posting, alongside the year-end journals and reconciliations. Identify what must happen first, what can safely overlap, and what depends on an overnight job finishing on time. If the sequence exists only in someone’s memory, or in a spreadsheet of workarounds, the process is already carrying key-person risk.
A professional services firm may have a compact finance team where the same person validates payroll and prepares close adjustments. A logistics business may rely on overnight batch processing to reconcile operational activity before finance can post. In either case, a delay in one run can obstruct the other. Two deadlines, one system and no spare week is not a test of individual commitment. It is a test of the design around them.

The useful test of Disaster Recovery is therefore specific: if payroll has to be rerun during that same fortnight, which restore point would be used, and what happens to ledger work completed since then? Restoring the system may get the application running again. Recovering the position means knowing which payroll results, approvals, journals and reconciliations remain valid, what must be replayed, and how to avoid posting the same transaction twice.
A backup report marked “successful” is not proof that the business can recover a usable finance position. The backup needs to be tested, and the restore needs to include the data and dependencies the process actually relies on. Recovery time and recovery point objectives should be agreed in business language: how long can payroll be unavailable, and how much approved work can the business afford to lose? Those targets should reflect the consequences of the December run, not merely a technical default. A generic outage plan may say how to restore a server, yet say nothing about which payroll data is safe to rerun against journals already posted.
A restore is only free if it has been rehearsed before the week it matters. Test a scenario that reflects the real sequence: identify the backup point, restore the relevant environment, validate payroll and ledger records with finance, then record what would need to be repeated. The ICO’s guide to data security emphasises timely restoration of access to personal data and regular testing of security measures. That principle becomes concrete when a payroll run and a close share the same recovery window.
For SMEs still relying on an ageing on-premise finance system, December can expose how much process has accumulated around the software: manual reconciliations, undocumented workarounds and a spreadsheet held by one person. A strategic review may point towards Dynamics 365 Business Central as a supported, documented, cloud-hosted finance and ERP core, with repeatable month-end and year-end processes on a vendor-maintained platform. Microsoft describes Business Central online’s built-in backup, restore and continuity capabilities in its service overview. Cloud hosting does not remove the need for business-owned procedures, access controls or recovery planning. It does provide a more supportable foundation to design them around.
A Dynamics 365 Jump Start scoping conversation started in November can be realistic for planning a new-year go-live, subject to scope, readiness and delivery decisions. A Business Central migration started in December is not a sensible way to reduce risk during a live payroll and close period. Trying to migrate and run both deadlines in December is how SMEs create the very risk they were trying to remove. The Fractal IT Director approach is to set the strategy first, then bring in specialist channel partners for the execution and technical capacity the plan needs.
The same review must include the less visible foundations. On some legacy estates, IBM i, also known as AS/400, carries the least glamorous and most load-bearing December work: scheduled jobs, batch reconciliation, payroll or stock processing overnight, output queues and a job scheduler that may not have been reviewed since it was written. The risk is not the platform’s age alone. It sits in the batch schedule nobody owns, a job silently failing into a queue, or the one administrator who knows the run order being unavailable when finance arrives.
IBM i Management should establish a documented schedule, alerting when jobs fail, and clear accountability for an overnight run that does not finish before the payroll team starts. Managed cover is particularly valuable where one person holds the operational knowledge. David’s practitioner-led perspective, shaped by Google-certified IT project management, multi-national onboarding and IBM i system administration, informs the strategic assessment; specialist delivery can then be scaled through trusted partners.

November’s spend review also has a December purpose. Technology Expense Management (TEM) is not cost-cutting for its own sake. It is an evidence-based way to find spend that could fund resilience work: extra licences bought for a temporary finance hire, duplicated payroll or HR subscriptions, a support contract that excludes out-of-hours help during payroll week, or temporary capacity that could have been planned and contracted rather than purchased in a panic. The point is to connect each cost to business need, contractual coverage and operational risk, then redirect avoidable spend towards the controls that matter. Evestaff’s group services gateway is relevant where wider business services form part of an organisation’s operational planning.
ISO 27001 and Cyber Essentials are consultancy domains, not products that solve this problem by themselves. December is a poor time to discover an overdue patch or an expired certificate. In the new year, an assessor may want to see evidence that tells a coherent story across the period: patching records, restore-test logs, incident records, temporary-staff access changes and a documented continuity arrangement for year-end. The NCSC’s Cyber Essentials guidance sets out its core technical controls, including security updates and user access control. The right advice is to understand how those controls apply to the business and its actual systems, then keep evidence that they operated in practice.
Payroll data is among the most sensitive personal data an SME holds. Access control and logging around the December run are precisely the sorts of controls an assessment may sample. Temporary access should be authorised, limited to what the role requires, recorded and removed when no longer needed. A payroll file sent through an informal route can undermine otherwise sound platform controls. The technology and the working procedure have to agree.
A strategic IT plan should clarify ownership, dependencies and recovery before the critical fortnight begins. Helpdesk Support and Managed IT Services can provide responsive operational cover; Network Solutions and Voice and Connectivity can support the infrastructure people rely on to complete approvals and coordinate work. These are bolt-on execution services behind the strategy, not substitutes for it. The right mix depends on which systems are critical, where the failure points sit and what recovery the business actually needs.
By mid-November, make the plan tangible. Confirm the payroll and year-end sequence, and name the person accountable for each step. Confirm which restore point would be used if payroll had to be rerun, and how ledger work from that week would be reconciled. Walk one runbook step for real, rather than simply confirming that a document exists. Confirm who is on call during payroll week. Then check the one thing everybody assumes somebody else checked.
December is not uniquely dangerous. It is uniquely unforgiving: when the first fortnight breaks something, there is no second fortnight available to fix it before the year closes.
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
Join The Discussion