AGSA Predictive Model

The AGSA general report is remarkably consistent year to year: the same focus areas — leadership, financial management, performance, compliance, asset management, SCM and internal control — generate the same repeat findings across the sector. That consistency is a gift to a CAE who can read it, because it means next year's findings are partly predictable from this year's repeat rate, overdue remediation and critical-severity history. Risk-based planning is supposed to exploit exactly this.

But most internal-audit plans are built on professional judgement applied to a backlog of memory, not on a structured read of the entity's own finding history. High-probability repeat areas get the same coverage as quiet ones, audit resources are spread by habit, and the function cannot show the Audit Committee why it weighted the plan the way it did. The predictable finding recurs because nobody pointed the plan at it.

A heuristic predictive model estimates the probability of a repeat AGSA finding next year per focus area — Low, Medium, High, Very High — from your own repeat-rate, overdue-remediation and critical-severity inputs. The formula is published and reproducible; nothing leaves the tenant; and the CAE weights the audit plan to the highest-probability areas with a defensible, auditable basis.