SynalogicInsights
Insights | AI Audit Guide Continuous Assurance AusTrac Tranche 2 AI Platform Comparisons

The eleven-month problem in internal audit

Annual audit engagements are snapshots. Between those snapshots, controls drift, access rights accumulate, and transactions fall outside parameters — for months before anyone notices. Continuous assurance closes that gap. This guide explains how.

The gap no-one wants to talk about

A typical internal audit engagement takes four to eight weeks. It examines controls at a point in time, produces a report, and generates a list of actions. Then the next annual cycle begins — and for the eleven or so months in between, the organisation's control environment runs without structured oversight.

Regulators, boards, and audit committees are increasingly asking not just what the last audit found, but what is happening right now. A finding that surfaces in an annual engagement almost always reflects a problem that existed months earlier. Continuous assurance addresses that — not by replacing formal audit, but by providing the monitoring layer that makes those engagements more effective and gives leadership real-time visibility.

11 moAverage gap between audit engagements
1–5%Data coverage in sample-based audit
100%Population coverage with continuous monitoring

Synalogic Sentinel monitors your control environment continuously — every alert documented, every resolution recorded.

What continuous assurance actually involves

Continuous assurance is not a single technology. It is a monitoring programme that runs in parallel with formal audit cycles, testing controls and conditions on an ongoing basis. A well-designed programme operates on two levels:

Tier 1 — Rules-based continuous checks

Defined thresholds, policy parameters, and compliance conditions tested automatically against 100% of your data continuously. Segregation of duties violations. Approval threshold breaches. Policy expiry. Mandatory field completeness. Deterministic — when a condition is met, a flag is raised. Transparent, auditable, explainable to any regulator.

Tier 2 — AI-assisted periodic analysis

Scheduled analysis that identifies patterns across the evidence base that static rule sets miss. AI is most valuable here for correlation — spotting that three individually unremarkable signals in different systems are better understood together. Draft findings with severity ratings and citations, reviewed by humans before anything is recorded.

Rules-based testing is the safe foundation because the logic is transparent — if a regulator asks why something was flagged or not flagged, there is a documented answer. AI adds coverage that rules alone cannot provide, but always with human review before any conclusion is recorded.

Why AI-only continuous monitoring creates a different kind of risk

Several platforms position themselves as AI-driven continuous monitoring — the AI determines what to flag, how to score it, and in some cases how to resolve it. The efficiency case is real. The accountability exposure is also real: when AI makes autonomous monitoring decisions, the organisation cannot fully explain its control environment to a regulator, and the human review layer that professional standards require is absent or inconsistently applied.

AI-only continuous monitoring

  • AI determines what to flag — auditor sees output, not the logic behind it
  • No mandatory human review gate on each alert
  • Difficult to explain to regulators why something was not flagged
  • Audit trail documents system activity, not professional judgment
  • Alert resolution by AI or optional human approval

Synalogic Sentinel

  • Rules-based Tier 1 testing — fully transparent, auditable logic
  • AI-assisted Tier 2 signal correlation — human reviews every output
  • Mandatory human review gate on every alert before resolution
  • Reviewer identity, decision, and timestamp logged permanently
  • Board-ready reporting from validated signals only

What Sentinel monitors

Every alert requires documented human review. Sentinel does not resolve alerts autonomously. Every signal flows into a human review queue. The reviewer, their decision, and the timestamp are logged permanently and exportable for board reporting, regulatory examination, or audit committee review.

Sentinel and Assure share the same platform. Signals from continuous monitoring flow directly into formal audit engagements. Evidence collected once. The same accountability framework throughout.

Continuous monitoring and formal audit — the complete picture

The most effective audit functions combine both. Continuous monitoring provides the breadth — testing 100% of the data population at all times. Formal audit engagements provide the depth — structured methodology, professional findings, defensible reports. Each strengthens the other.

When Sentinel and Assure operate on the same platform, this integration is concrete rather than conceptual. Signals Sentinel surfaces flow directly into Assure engagements as pre-collected evidence. The audit team does not need to re-request or re-examine material that continuous monitoring already documented. Engagements start with a richer evidence base and a clearer picture of where professional effort is most warranted.

Common questions

Questions from audit professionals considering continuous assurance programmes.

What is continuous assurance and how is it different from internal audit?
Internal audit is a periodic structured engagement producing a formal report with documented findings. Continuous assurance is the ongoing monitoring of controls, transactions, and compliance obligations between those engagements. The two are complementary: audit provides the depth of a formal professional engagement; continuous assurance provides the breadth of constant oversight that catches issues before they become reportable findings. Synalogic Sentinel covers continuous assurance; Synalogic Assure covers internal audit engagements.
What is the best continuous monitoring software for internal audit?
The best continuous monitoring software combines rules-based control testing with AI-assisted signal correlation and mandatory human review. Synalogic Sentinel uses Tier 1 rules-based continuous checks, Tier 2 AI-assisted analysis, and a mandatory human review gate on every alert — with the full signal history exportable for regulatory or board reporting.
Why is there audit risk between engagements?
Traditional audit engagements are snapshots. Between those snapshots, controls drift, access rights accumulate, policy compliance lapses, and transactions fall outside expected parameters without anyone noticing. For organisations with annual audit cycles, this means up to eleven months of unmonitored exposure. Continuous monitoring closes that gap by testing controls on an ongoing basis, giving audit teams and boards real-time visibility into the control environment.
Should continuous monitoring use AI or rules-based testing?
Both in combination, with rules-based testing as the foundation. Rules-based tests have defined transparent logic: they flag what you have told them to flag, so you can explain every alert to a regulator or board. AI-assisted analysis adds coverage by correlating signals across large populations and identifying anomalies outside pre-defined patterns. Synalogic Sentinel uses rules-based Tier 1 testing and AI-assisted Tier 2 analysis, with human review required on both.
How does Synalogic Sentinel work with Synalogic Assure?
Sentinel and Assure share the same validation architecture and audit trail standard. Signals from Sentinel flow directly into Assure engagements — evidence collected continuously does not need to be recollected during a formal audit. The combination gives audit functions both the depth of a formal engagement and the breadth of continuous oversight on the same platform, with the same accountability framework throughout.

Close the gap between your audit engagements

See how Synalogic Sentinel monitors your control environment continuously — every alert reviewed and documented, findings ready when your next formal audit begins.