Demonstration environment. Fictional organisations and data where not signed in to your workspace.

Flagship module, Internal controls

AI SOX Engine

Maps risks to controls, tests populations rather than samples, and surfaces design and operating deficiencies with evidence attached - every conclusion routed to a named control owner for sign-off.

Halden Group plcFY26 · Q3 testing cyclePCAOB AS 2201 aligned

Loading…

Demonstration

SOX testing workflow

From population extraction to owner sign-off. Every hand-off is governed.

1System
Ingest populations
Journal, access and configuration data extracted read-only from ERP and GRC.
2AI-supported
Map risks to controls
Risk & control matrix reconciled to process narratives.
3AI-supported
Test full populations
Deterministic tests run across 100% of transactions.
4AI-supported
Evaluate deficiencies
Likelihood × magnitude assessment with explainability record.
5Human
Owner review & sign-off
Named control owner approves, edits or rejects each conclusion.

Run the workflow to see how automated steps hand off to accountable reviewers.

Optimisation scenario

Modelled impact of control automation ahead of first 404(b) attestation

SCN-SOX-AUTO · Control optimisation

Automate 40% of key controls

Convert manual detective controls in R2R and P2P to automated preventive controls ahead of the first SOX 404(b) attestation.

Modelled

Assumptions

Controls converted: 164 of 412
Implementation window: 2 quarters
ERP changes: Workflow + GRC rules

Modelled outcomes

Testing hours

-6,200 hrs

-38%

Deficiency likelihood

-27%

Favourable

One-off cost

£1.4m

Neutral

Model confidence83% · Moderate

Owner: Daniel Okafor, Head of Internal Audit · Updated 2 Oct 2026

AI-supported output. Human review and approval required before use.