IT operations controls: job scheduling, interfaces, backups and incident management

How to test operations controls that keep batch jobs and interfaces complete, protect data through backups, and resolve incidents affecting financial data.

4 min read · CORAA Team
Lesson 7 of 14 · 0 of 14 complete
In this lesson you will
  • Identify batch jobs and interfaces that feed financial reporting
  • Test job monitoring and failure resolution controls
  • Evaluate backup and restore controls in proportion to financial reporting risk
  • Test incident management for issues that affect financial data

What operations controls protect

AS 2201 names computer operations alongside program changes and access as the general controls that must be effective for benchmarking automated controls. The financial reporting risk is that processing does not happen, happens twice, or happens incompletely, and nobody notices. A nightly interface that posts sales to the ledger is an example: if it fails for two days and is rerun without checks, revenue could be missing or duplicated.

Job scheduling and interfaces

Start by listing the scheduled jobs and interfaces that move or transform financial data for in-scope applications: sales feeds, payroll postings, bank statement imports, depreciation runs, intercompany allocations.

Key controls:

  • Changes to the job schedule are authorized (this overlaps with change management).
  • Job failures generate an alert or exception report.
  • Someone reviews failures and resolves them, with evidence.
  • Interfaces include completeness checks (record counts or control totals compared between source and destination).

Testing approach:

  1. Obtain the job failure log for the period from the scheduling tool.
  2. Select a sample of failures and trace each to resolution: ticket, rerun evidence, and confirmation that the rerun completed without duplicates.
  3. For interface reconciliations, inspect a sample of daily or monthly reconciliations of record counts or totals, and follow up differences.
  4. Inquire and inspect who can change the job schedule; restrict to authorized operators.

Backups and restores

Backups protect data from loss. For financial reporting, the relevant question is whether the entity could lose financial data in a way that leads to misstatement. Key controls:

  • Backups of in-scope databases run on schedule, and failures are followed up.
  • Restores are tested periodically to show backups are usable.

Testing approach: inspect the backup schedule and a sample of backup logs, follow up failures, and inspect evidence of at least one restore test. A backup failure without data loss does not misstate anything, so backup controls carry less ICFR risk than access and change controls in most environments. Document your rationale either way.

Incident management

Incidents include processing errors, data corruption, and security events. The control is that incidents affecting financial systems are logged, prioritized, investigated and closed, with root cause addressed.

Testing approach: obtain the incident log for in-scope systems, filter for incidents touching financial data, and inspect a sample for resolution. Read the descriptions; an incident like "duplicate invoices posted on March 14" is not just an IT matter. It is a potential misstatement that needs follow-up in the substantive audit.

Security events

A security incident such as unauthorized access or ransomware may affect the integrity of financial data. Ask management about security incidents during the period and how it determined whether financial data was affected. Coordinate with the audit team on any effect on the financial statements and disclosures.

Cloud and hosted systems

When a service organization hosts the application, job scheduling, backups and incident management usually sit with the provider. The evidence comes from the SOC 1 report, and the client often has complementary user entity controls such as reviewing interface reconciliations. Lessons 11 and 12 cover how to use that report.

Proportion

Operations controls are often less likely to cause material misstatement than access or change controls, but interface completeness is an exception: a broken interface misstates revenue or expenses directly. Spend the time where data moves into the ledger.

Worked example

Interface failures at Silver Fern Retail

Silver Fern Retail's point-of-sale system sends daily sales summaries to the general ledger through an overnight interface. The accounting team reconciles store sales totals to ledger postings each morning.

The scheduling tool log shows 14 interface failures during the year. The team selects all 14 because the population is small.

Results: 12 failures were rerun the same day, with the reconciliation showing agreement after the rerun. On two dates in November, the interface was rerun twice by different operators, and the morning reconciliation was skipped because the accountant was on leave.

The team compares ledger postings for those two dates to point-of-sale reports: $412,000 of sales was posted twice.

Management had already found the duplication during the November close review and reversed it before the month-end close. The team inspects the reversing entry, dated November 30, and agrees it to the duplicate postings.

Evaluation: the interface monitoring control failed on two dates, and the morning reconciliation was not performed. A detective control (the monthly close review) caught the error. The team tests the precision of the monthly review and concludes it would detect a duplicate posting of that size.

Conclusion: an operating deficiency in the daily reconciliation control, mitigated by the monthly review. No unadjusted misstatement. The team recommends a backup reviewer for the daily reconciliation.

Self-check

Answer each question before you open the answer.

  1. Which operations control most directly affects financial statement amounts?

    Show answer

    Interface and job completeness controls, because a failed or duplicated interface posts too little or too much to the ledger. Backups matter mainly when data is actually lost.

  2. What should you do when an incident log shows duplicate postings?

    Show answer

    Treat it as a potential misstatement as well as an IT control matter: determine whether it was corrected, inspect the correcting entries and evaluate the related control failure.

  3. Where does evidence about operations controls come from when the application is hosted by a service organization?

    Show answer

    Usually from the service organization’s SOC 1 Type 2 report, together with the user entity’s complementary controls such as interface reconciliations.

Sources

Back to the syllabus

Standards are paraphrased from the primary sources listed above; read the source before relying on a requirement. This lesson is training material, not professional advice. Report an error

For US CPA firms

Join the CORAA waitlist

CORAA AI drafts audit working papers from your client’s QuickBooks books and does the first pass: lead schedules, tie-out, memos, draft notes and letters. CORAA Network reviewers clear every finding. Your firm reviews and signs.

  • First audit free, so you can judge the quality on a real file
  • Flat fee by entity size; messy books included
  • We work only with CPA firms, never their clients
See how it works →

* required