L Lumina Risk Advisory

[ Blog ] SOX · ITGC · Pre-IPO

The SOX ITGC checklist
every pre-IPO company
needs in Year 1.

March 2026  ·  10 min read  ·  Lumina Risk Advisory

Year 1 of SOX is the hardest audit your IT team will face. Your external auditors are scoping everything for the first time. Your controls may have existed informally for years but never been documented. And your evidence library doesn't exist yet. This checklist gives you the starting framework.

Before you use this checklist

This is a directional starting framework — not a substitute for a scoping engagement. Your actual ITGC scope depends on your in-scope systems, your auditor's risk assessment, and your company's specific technology stack. Use this to understand the shape of the work; use a proper scoping engagement to define the actual scope.

What are ITGCs, and why do they matter for SOX?

IT General Controls (ITGCs) are the foundational controls that govern how systems are accessed, changed, operated, and developed. They are "general" because they apply to all applications — unlike application controls, which are specific to a single system.

Under SOX Section 404, management must assess the design and operating effectiveness of internal controls over financial reporting (ICFR). ITGCs underpin that assessment: if your ITGCs are weak, your financial application controls can't be relied upon — which means your entire ICFR assessment is at risk.

External auditors test ITGCs because they need to confirm that the systems producing financial data are reliable. If access controls are weak, someone could have manipulated financial data. If change management controls are weak, unauthorized changes could have introduced errors. The auditors need comfort on both fronts before they'll sign off on the financial statements.

The four ITGC domains — with evidence requirements

Your auditors will test across all four of these domains. For each control, we've listed the control objective, what evidence auditors will ask for, and what risk it addresses.

AC

Access to Programs & Data

User provisioning process

Risk: Unauthorized access to financial systems

Evidence auditors will request

IT tickets or access request forms showing approval before access granted

User de-provisioning process

Risk: Former employees retaining system access

Evidence auditors will request

Terminated user list cross-referenced to system access removal within defined SLA

Privileged access management

Risk: Excessive privilege creating segregation of duties violations

Evidence auditors will request

List of admin/privileged users with documented approval; quarterly access review sign-off

Periodic access reviews

Risk: Accumulated access over time not reviewed or recertified

Evidence auditors will request

Signed access review certifications by system owners at least quarterly

Segregation of duties

Risk: One person controlling an entire financial process

Evidence auditors will request

Role matrix showing no individual has conflicting access (e.g., create and approve journal entries)

Authentication controls

Risk: Weak authentication enabling unauthorized access

Evidence auditors will request

MFA enrollment evidence, password policy configuration screenshots

CM

Change Management

Change request and approval process

Risk: Unauthorized changes to financial reporting systems or data

Evidence auditors will request

Ticketing system showing change requested, tested, approved by separate approver, and deployed

Segregation between development and production

Risk: Untested changes deployed directly to production

Evidence auditors will request

Evidence that developers cannot directly deploy to production; CI/CD pipeline controls or access matrix

Testing requirements

Risk: Untested changes causing system errors in financial processing

Evidence auditors will request

Test plan and results documented for significant changes before production deployment

Emergency change procedures

Risk: Emergency changes circumventing standard controls without compensating review

Evidence auditors will request

Documented emergency change process; examples showing retroactive approval within defined timeframe

CO

Computer Operations

Job scheduling and monitoring

Risk: Batch processing failures going undetected and affecting financial balances

Evidence auditors will request

Batch job schedules, automated alerts, exception reports reviewed and signed off

Backup and recovery

Risk: Data loss due to system failure with no recovery capability

Evidence auditors will request

Backup schedule, backup completion logs, periodic recovery test results

Incident management

Risk: System incidents affecting financial reporting not tracked or resolved timely

Evidence auditors will request

Incident log showing issues detected, escalated, resolved within SLA; post-incident reviews for material incidents

System monitoring

Risk: Performance or availability issues not detected until they affect financial processing

Evidence auditors will request

Monitoring tool configuration, alert thresholds, evidence of alerts reviewed

SD

System Development (SDLC)

SDLC methodology

Risk: New systems implemented without adequate controls or testing

Evidence auditors will request

Documented development process showing requirements, design, testing, and deployment phases

User acceptance testing

Risk: Systems placed in production that don't meet requirements or contain errors

Evidence auditors will request

UAT sign-off from business owners before new systems or significant changes go live

Post-implementation review

Risk: Implementation issues not identified and corrected after go-live

Evidence auditors will request

Evidence of review conducted after significant implementations to confirm expected outcomes

The most common Year 1 findings

In our experience, these are the findings that show up most often in first-year SOX IT audits:

  • Access reviews not completed on schedule or lacking documented sign-off from system owners
  • Terminated employees' access not removed within the policy-defined timeframe (often set too generously at "within 30 days")
  • Developers with direct production access — common at pre-IPO companies where the engineering team hasn't yet been segregated
  • Change management tickets missing approver, missing test evidence, or closed without documentation of what was deployed
  • No formal user acceptance testing process — changes tested informally but not documented
  • Backup testing not performed or not documented — backup processes exist but recovery has never been tested
  • Admin/privileged accounts not formally approved and reviewed quarterly

What "significant deficiency" and "material weakness" actually mean

Not all audit findings are equal. External auditors classify IT control deficiencies on a spectrum:

  • Control deficiency A weakness in the design or operation of a control that could affect financial reporting but is unlikely to be material. Common in Year 1. Noted in the management letter but not in the audit opinion.
  • Significant deficiency A deficiency, or combination of deficiencies, that is less severe than a material weakness but important enough to merit attention by those responsible for oversight. Disclosed in the SOX 404 report.
  • Material weakness A deficiency that creates a reasonable possibility that a material misstatement of financial statements will not be prevented or detected on a timely basis. Must be disclosed in your 10-K. Sends a signal to investors and auditors. This is the one to avoid.

The best defense against all three is remediating before the audit window opens — not after the auditor finds it.

Need help scoping and prioritizing your Year 1 ITGC work? A scoping call will tell you where to focus first and what the audit window looks like from here.