Safeguard
Compliance•

SOX IT general controls, a 12 week implementation calendar

A week by week plan for SOX IT general controls covering scope, access, change management, operations and testing, with the evidence to save each week.

Safeguard Security Team
5 min read

You can stand up SOX IT general controls in about twelve weeks if scope is clear and one person owns each domain. Spend weeks one to three on scope and risk, four to nine on access, change and operations controls, and the last three on testing and fixing. The twelve weeks are a planning assumption, not a requirement.

What are IT general controls?

IT general controls, or ITGCs, are the controls over the systems that produce or protect financial reporting data. They are usually grouped into access to programs and data, program change management, and computer operations such as backup and job scheduling. They matter because management and external auditors rely on automated controls and reports in those systems, so a weak ITGC puts that reliance in doubt.

Source note: the SEC and PCAOB pages we tried to fetch returned errors, so this guide describes common practice and not quoted text. Confirm scope, testing approach and deficiency evaluation with your external auditor and finance leadership. The calendar below is our planning suggestion.

What does each week look like?

Weeks 1 to 2, scope. Finance lists the financially significant applications, such as the general ledger, billing and payroll, plus the databases, operating systems and infrastructure under them. IT adds the supporting identity provider and deployment tools. Output: a system list with owners and the reason each is in scope.

Week 3, risk and control design. For each system, write the risks in the three domains and draft a control for each. Keep the list short. Ten well run controls beat forty unowned ones. Output: a risk and control matrix.

Weeks 4 to 5, access. Define the request, approval, provisioning and removal workflow. Review who has privileged access, remove shared admin accounts, and enforce multi-factor authentication. Set a quarterly user access review for each in-scope system. Output: first completed review with sign off and removals.

Weeks 6 to 7, change management. Require tickets, peer review, testing evidence and approval before production changes. Separate people who write code from those who release it, or add a compensating review. Define the emergency change path with after the fact approval. Output: a sample of recent changes traced end to end.

Week 8, operations. Document backups, job monitoring and incident handling for in-scope systems. Run a restore test. Output: restore record and a job failure log with follow up.

Week 9, evidence design. For every control, define the evidence, how often it is produced, who reviews it and where it is stored. Population lists, such as all changes in a period, must be reproducible from the system, not rebuilt from memory.

Weeks 10 to 11, dry run testing. Someone independent of the control owner samples each control the way an auditor would and records exceptions. Output: deficiency list.

Week 12, remediate and brief. Fix what failed, document what you changed, and brief finance leadership and the audit committee on remaining gaps.

Who does what?

The CFO or controller owns the scope decision and the link to financial reporting. The head of IT or engineering owns control operation. A compliance or internal audit lead owns testing independence. Without one named owner per control, reviews are skipped.

What mistakes appear most?

  • Scoping by system name rather than by whether it supports a financial process.
  • Access reviews where the reviewer approves everything in bulk.
  • Change records written after deployment.
  • Evidence that depends on a screenshot and cannot show the full population.
  • Treating a service provider as out of scope without reading their assurance report.

What to do

  1. Agree the in-scope systems with finance and the external auditor.
  2. Build the risk and control matrix across access, change and operations.
  3. Complete one access review and one restore test early.
  4. Make change approval a system enforced gate.
  5. Define evidence and populations per control.
  6. Run an independent dry run before the audit window.

Quick check

What are the usual ITGC domains?

Access to programs and data, program change management and computer operations are the common groupings.

Is twelve weeks mandatory?

No. It is a planning assumption. Real timing depends on scope, staffing and what you already have.

How Safeguard helps automate this

Safeguard's compliance suite covers frameworks, controls, evidence, continuous monitoring, automated tests and questionnaires. Evidence for change and vulnerability controls can be drawn from connected systems and your real code and dependencies through scans, SBOMs and reachability, which also gives a reproducible population for sampling. Monitoring runs between audit windows, which frees the team for other work. Safeguard's agent can open tested fix pull requests for vulnerable dependencies. It is off by default, every category starts on Manual, and you choose which categories run automatically and the approval level. Your external auditor forms the opinion. Safeguard helps you prepare and keep evidence ready. Book a call to see it.

Sources

Last reviewed: 18 October 2026.

Never miss an update

Weekly insights on software supply chain security, delivered to your inbox.

Self-healing security runs on Safeguard.

Your first fix PR is minutes away.

No sales call required, even your agent can complete the purchase over MCP.