What is an AI Readiness Audit? (and what you actually get)

An AI Readiness Audit is a short, fixed-fee diagnostic that tells you — before you spend a dollar on building anything — whether your organisation can safely and profitably deploy AI, where the highest-value opportunities are, and what has to be fixed first. For the way we run it, that means two weeks, senior-led, ending in a small set of decision-grade artefacts you can put in front of a board.

That is the one-paragraph answer. The rest of this guide explains what it assesses, what you actually receive, and — just as importantly — when you do not need one.

What does an AI Readiness Audit assess?

Most AI projects that fail in Australian mid-market firms do not fail because the model was wrong. They fail because the data was not ready, the governance was not in place, or the chosen use case was never going to pay back. An audit looks at the things that actually determine success:

• Data — what you hold, where it lives, how clean it is, and whether it can feed an AI system without a six-month plumbing project first.
• Risk and regulation — the regulatory shape of each workflow: Privacy Act 2026 obligations, APRA CPS 230 where relevant, the Voluntary AI Safety Standard, and whether automated decisions need to be explainable.
• Use cases — the candidate processes, scored on value, feasibility and risk, so you start with the one most likely to pay back rather than the one that demos well.
• Governance — the gap between how you make decisions today and what you would need to run an AI system you can defend.

What do you actually get?

Not a slide deck of generic AI trends. The audit produces four artefacts that are useful even if the recommendation is "fix your data first":

Deliverable	What it's for
Readiness scorecard	An honest, evidence-based rating of your data, governance and risk posture.
Prioritised use-case shortlist	The highest-value workflows, scored on value vs. feasibility vs. risk — so the first build is the right one.
Governance gap analysis	The specific controls you need before deploying, mapped to the relevant AU regulations.
R&D Tax Incentive eligibility memo	Whether the work qualifies for the R&DTI — often material to the business case.

These are designed to stand on their own. The scorecard and governance analysis are valuable artefacts whether or not you build anything next.

How long does it take, and what does it cost?

Two weeks from kickoff. It is fixed-fee — you know the number before you start, and it does not move. We publish the timeline, not the price, because the fee is scoped to your context (data maturity, governance posture, the workflows in question) and we share it on the discovery call once there is enough to scope it accurately.

It is delivered by senior practitioners — the audit is not handed to a graduate to run. If we cannot staff it senior, in Australia, we do not take it.

When do you actually need one?

You are a good candidate for an AI Readiness Audit if:

• You are a mid-market firm with real, document-heavy or rule-bound processes.
• Leadership is being asked "what's our AI plan?" and you want an evidence-based answer, not a vendor's.
• You have a workflow you suspect AI could materially improve but you are not sure it is safe or worth it.
• You operate in a regulated industry and need automated decisions to be auditable.

You probably do not need one if your problem is small enough to solve with an off-the-shelf tool, or if you have already done a rigorous internal assessment and simply need delivery.

What if the audit says you're not ready?

Then it says so — in the readout. "Fix your data first" is a perfectly common and genuinely useful outcome. The cost of finding that out in two weeks is trivial next to the cost of discovering it nine months into a stalled build. The artefacts still give you a concrete plan: what to fix, in what order, and what becomes possible once you do.

How is this different from a vendor's "free AI assessment"?

A vendor's assessment is a sales motion — it tends to conclude that you need the vendor's product. An independent audit has no product to sell you at the end of it. The recommendation might be to build, to fix data first, to buy something off the shelf, or to do nothing yet. The point is a defensible decision, not a foregone one.

---

If you want to know whether AI is worth it for a specific process in your business — and what it would take to do it safely — that is exactly what the audit is for. The next step is a free, 30-minute discovery call: we listen, ask sharp questions, and tell you whether the audit is your right next step, or not.