BusinessTech

When the ERP Goes Down: A Financial Controller’s Month-End Close Continuity Playbook

A practical finance fallback plan for ERP outages: prioritize critical close activities, preserve approvals, track manual transactions, protect payment controls, and reconcile everything after recovery.

Month-end does not pause when an ERP, consolidation platform, bank portal, or reporting interface becomes unavailable. A prolonged outage can delay journal entries, reconciliations, cash visibility, approvals, and management reporting—and can increase the risk of unapproved spreadsheet workarounds or duplicate postings after recovery.

Finance needs a controlled fallback, not an improvised folder of spreadsheets. The aim is to maintain essential accounting processes for a limited period, protect evidence and segregation of duties, and return to the normal system without losing transactions or creating duplicate entries. This practical playbook is informed by NIST contingency-planning guidance and the Federal Reserve’s operational-resilience discussion. NIST notes that short-term manual alternate processing can be one contingency approach; its guide emphasizes business-impact analysis, recovery priorities, testing, training, and plan maintenance. These are useful design principles for finance, not a claim that the federal guidance is a mandatory rule for every private company.

1. Define the close processes that must continue

Do not attempt to replicate the entire ERP manually. Identify the activities that are time-critical, legally or contractually sensitive, or necessary for management decisions. Typical priorities might include cash position, urgent payment decisions, payroll support, revenue cutoff evidence, critical accruals, close communications, and preservation of source documents.

Rank activities into tiers—for example, essential same-day tasks, tasks that can wait 24–48 hours, and work that should stop until systems are restored. Assign a business owner to each process and define the trigger for activating the fallback. Consider dependencies: a payment process may rely on a bank portal, supplier master data, approval workflow, and cash forecast even if the ERP is unavailable.

2. Set realistic outage thresholds and escalation triggers

Use a business-impact assessment to establish when the outage becomes a finance disruption requiring formal escalation. Define expected recovery time and recovery point targets with IT and business owners, but do not treat a target as a guarantee. NIST’s contingency-planning material describes using business impact analysis to identify process criticality, outage impacts, resources, and recovery priorities.

Write down who can declare a finance fallback, who can authorize extension, who informs the CFO, and who decides whether the close calendar or external reporting timetable needs adjustment. Escalation triggers could include inability to access the ledger for a specified period, loss of critical data, unavailable approvals, or uncertainty about data integrity.

3. Prepare controlled manual workpapers before an outage

Build and test templates in advance rather than creating files during a crisis. Use approved formats for manual journal requests, cash reporting, AP/AR exception logs, accrual estimates, reconciliation status, transaction tracking, and management reporting. Each template should include a unique reference number, business date, entity, preparer, reviewer, supporting-document link or reference, approval status, and eventual ERP posting reference.

Keep templates in an approved location with access restricted to assigned personnel. Version-control the templates and designate a single master copy. Avoid emailing uncontrolled copies around the organization; competing versions can create duplicate or inconsistent transactions.

4. Preserve approvals and segregation of duties

System workflow may be unavailable, but approval requirements do not disappear. Define an alternate approval channel, such as a restricted shared workspace or documented email approval, and identify who may approve transactions if normal approvers are unavailable. Separate preparation, review, and release wherever practicable. If a temporary role combination is unavoidable, document it, limit the duration, and require independent retrospective review.

For payments, maintain the normal authorization limits and bank-release requirements. Do not use the outage as a reason to change vendor bank details through an unverified channel or to bypass callback and dual-approval safeguards.

5. Establish a manual transaction log

Every finance activity performed outside the system during the disruption must be captured in a centralized log. Assign a unique transaction ID before work proceeds. Record the transaction type, entity, date, amount and currency, supporting evidence, preparer, reviewer, approval, status, and the system document number once entered after recovery.

Use status values that are easy to reconcile, such as submitted, reviewed, approved, posted, rejected, and reconciled. The log should allow the controller to identify items that are approved but not posted, posted but not matched to the original request, or duplicated in the system after restoration.

6. Keep critical activities small and risk-based

Use the fallback only for defined priorities and maintain a queue for deferred tasks. For example:

  • Cash: maintain a controlled cash-position report using verified bank data; clearly timestamp the source.
  • Payments: process only critical payments under normal authorization limits and bank security procedures.
  • Close: capture accruals and cutoff evidence in approved templates; flag estimates that require later reassessment.
  • Revenue and shipments: retain source evidence and log items requiring subsequent system entry; do not make unsupported manual recognition decisions.
  • Reconciliations: update status and record known differences, but label interim work clearly until system data is restored.
  • Reporting: communicate the limitations, date/time of information, missing data, and whether reports are preliminary.

7. Communicate one status, one owner, and one set of priorities

Finance should receive a brief scheduled outage update from a designated coordinator. It should state what is unavailable, the approved workaround, tasks to defer, current escalation contact, and the next update time. Department leaders should not independently create new accounting procedures during the outage.

For multi-entity organizations, appoint local finance contacts who report into a central coordinator. Use one shared status tracker to identify affected entities, critical transactions, overdue approvals, and tasks that cannot proceed. Keep messages concise and distinguish confirmed facts from estimates.

8. Plan the recovery and reconciliation before using the workaround

Before fallback processing starts, decide how transactions will be re-entered or imported after service returns. Assign the reconciliation owner and define the matching key: unique manual ID, document number, amount, entity, date, and transaction type. If system data must be restored from backup or replayed from interfaces, confirm with IT how integrity and completeness will be assessed.

Do not simply upload or rekey the entire manual log. Compare the log to the restored ERP, identify transactions already recorded by automated processes or recovered data, and mark each item as posted, duplicate, rejected, or pending. Reconcile control totals by type and entity and investigate differences.

9. Retain an audit trail and conduct a post-outage review

Preserve the outage declaration, communications, workpapers, approvals, source evidence, manual log, posting references, reconciliation, and any temporary access changes. Record who performed and reviewed each action. When normal processing resumes, confirm temporary permissions and workarounds are removed or formally extended.

Within a reasonable period, review what worked and what failed: templates, access, approvals, recovery timing, data completeness, bottlenecks, and staff training. Update the plan and rehearse it. NIST’s contingency-planning process explicitly includes testing, training, exercises, and plan maintenance, not just writing a plan.

Controller’s ERP-outage checklist

  • Finance fallback activation trigger and escalation roles are defined.
  • Critical and deferrable close processes are ranked.
  • Approved manual templates and controlled storage are ready.
  • Alternate approval channel preserves limits and separation of duties.
  • Every workaround transaction receives a unique reference and central log entry.
  • Payments and vendor changes retain independent verification safeguards.
  • Interim reports are timestamped and clearly marked as preliminary when appropriate.
  • Recovery plan prevents duplicate posting and includes source-to-system reconciliation.
  • Temporary permissions are reviewed and removed after normal service returns.
  • Outage rehearsal, lessons learned, and plan updates are scheduled.

Conclusion

A finance outage plan should be deliberately limited, documented, and rehearsed. The controller’s job is to keep the most important activities moving without weakening authorization, data integrity, or auditability. A short manual log and disciplined recovery reconciliation are often more valuable than a complicated attempt to recreate the entire ERP in spreadsheets.

Informational only. Tailor the fallback to your systems, close processes, contractual obligations, security requirements, and reporting deadlines. NIST publications are guidance and do not automatically establish obligations for every organization.

Sources and further reading

Shares: