An AI readiness audit is a short, structured assessment of where AI creates value and risk in your specific workflows — covering data, skills, governance and quick wins — and in the UK it should cost £3,500–£15,000 and take one to three weeks. It's the first engagement most firms in our comparison of the best AI consulting firms in London and the UK will propose, which is exactly why you should know what a real one contains before anyone scopes yours.
Key Takeaways
- A real AI readiness audit assesses five things: workflow value mapping, data and stack readiness, team skills baseline, governance and risk posture, and a prioritised quick-win list. Missing two of the five means you're buying a partial audit at full price.
- UK price bands: free self-serve scorecards for orientation, £3,500–£15,000 for boutique audits with executive readout, £50,000+ for Big 4 versions with assurance-grade documentation.
- Duration is a tell: a proper audit takes days to weeks. A "12-week AI assessment" is a strategy phase wearing an audit's name tag — and priced accordingly.
- The most common failure is the audit-as-sales-document: a findings deck reverse-engineered to justify the vendor's build proposal. Independence checks below.
- The skills baseline is the most skipped component — and with over 60% of UK businesses citing the skills gap as their main AI barrier, it's the one that most changes what you do next.
What is an AI readiness audit?
An AI readiness audit is a fixed-scope diagnostic that maps where AI can create measurable value in your organisation's real workflows, what currently blocks that value — data, skills, governance or process — and which moves to make in the next quarter. It is not a strategy (that comes next), not a technical proof of concept, and not a maturity survey that scores you 2.7 out of 5 and stops. The output is decisions you can act on: this workflow first, this rule for data, this team trained first, this build deferred.
The five things a real audit covers
| Component | Questions it answers | What you should receive |
|---|---|---|
| 1. Workflow value map | Where do hours and errors actually concentrate? Which workflows are AI-suited? | Ranked use-case portfolio with effort/impact scores |
| 2. Data & stack readiness | Can current systems and data support the top use cases? What's licensed and unused? | Gap list per priority use case, licence utilisation snapshot |
| 3. Skills baseline | Who uses what today, at what depth? Where are the champions and sceptics? | Usage and confidence baseline per team |
| 4. Governance & risk posture | What rules exist? What does UK GDPR / ICO guidance / your sector require? | Risk register and policy gap list legal can review |
| 5. Quick wins & sequencing | What pays back inside 90 days? What order de-risks the rest? | 30/60/90-day recommendation with named owners |
Ask any proposing vendor to show which of the five their audit covers, in the deliverables, in writing. This single table filters most weak proposals — and it pairs with the ten harder questions in how to choose an AI consultancy.
What does an AI readiness audit cost in the UK?
- Free — self-serve scorecards. Orientation-level: ours takes 20 minutes and benchmarks you against UK adoption data. Right for deciding whether to spend anything at all.
- £3,500–£15,000 — boutique audits. Senior practitioners, one to three weeks, the five components above, executive readout. Ours is £3,500 including the executive strategy briefing, net of VAT.
- £50,000+ — Big 4 / large-firm assessments. Assurance-grade documentation, multi-stakeholder interviews at scale. Justified when the audit must survive regulatory or board scrutiny, oversized otherwise.
Fuller market bands, day rates and comparison tactics are in our UK AI consulting cost guide.
"The audits that worry me are the free ones attached to a build pipeline. If the auditor's business model is the £200,000 project the audit recommends, you didn't buy a diagnosis — you bought a brochure with your logo on it. Pay a small fixed fee to someone who doesn't build, and the findings change shape remarkably."
— Toni Dos Santos, Co-Founder, Spicy Advisory
Red flags in AI audit proposals
- The pre-written conclusion. The audit firm also sells the platform, integration or build the audit will inevitably recommend. Ask what percentage of their audits led to a recommendation someone else delivered.
- Tech-only scope. Data estate and architecture assessed; humans absent. If there's no skills baseline, the number one UK adoption barrier — cited by over 60% of businesses in the 2026 UK SME AI statistics — goes unmeasured.
- Survey theatre. A questionnaire produces a maturity score and a heatmap; nobody watched anyone work. Real audits sit with the teams doing the workflows.
- No sequencing. Forty use cases, unranked, is homework returned unmarked. The audit's job is the order, not the list.
- Open-ended duration. Weeks, fixed fee, named end date — or it's not an audit.
"Leaders consistently over-estimate their data problem and under-estimate their skills problem. In most audits we run, the licence utilisation number lands within a few points of 20%, and the room goes quiet. That one measurement usually redirects more budget than every architecture slide combined."
— Meera Sanghvi, Co-Founder, Spicy Advisory
After the audit: what comes next
An audit ends in decisions; the natural next step is a short strategy sprint that turns them into an AI charter, a governance framework and a 30/60/90-day roadmap — ours is two weeks, structured as described on our AI strategy consulting page. Then execution follows the diagnosis: adoption training if the gap is behaviour, a build partner if the gap is capability — the fork we've mapped in advisory vs implementation vs training.
Start free, honestly: our AI Adoption Scorecard is the orientation tier of this exact process — 20 minutes, benchmarked against current UK data, no sales call required to see your results. If the scorecard says you don't need the paid audit yet, believe it.
Frequently asked questions
How long should an AI readiness audit take?
One to three weeks for a 50–1,000 person organisation: enough to interview teams, measure the skills and usage baseline, and rank use cases. Multi-month "audits" are strategy phases priced under a different name.
How much does an AI readiness audit cost in London?
The same as the rest of the UK in practice: free at scorecard level, £3,500–£15,000 from boutiques with senior delivery, £50,000+ from the Big 4. Remote delivery has removed most of the London premium at audit scale.
What should an AI readiness audit deliver?
Five artefacts: a ranked use-case portfolio, a data and licence gap list, a per-team skills baseline, a governance risk register, and a sequenced 30/60/90-day recommendation with owners. A maturity score alone is not a deliverable.
Who should run the audit — internal team or consultancy?
Internal teams know the workflows; outsiders get honest answers and bring cross-company baselines. The cheapest robust option is usually a small external audit with named internal counterparts, so the method transfers and the next audit can be yours.
Sources
- British Chambers of Commerce / Atos and ONS Business Insights, 2026 — via our UK SME AI statistics roundup
- ICO — AI and UK GDPR guidance — the governance baseline UK audits should map against
The £3,500 version, or the free version
We run readiness audits as a fixed-scope engagement: five components, senior-only delivery, executive briefing included, £3,500 net of VAT — and we don't sell builds, so the findings have nowhere to hide. Not ready to spend? The scorecard is genuinely free.
Run the Free AI Adoption Scorecard Book a Free 20-Minute Call