Entry · 3 working days · £397

You've already tried AI. We'll tell you why it isn't working.

A diagnostic review of your existing AI setup - Claude Project, Custom GPT, Jasper, whatever you've built. We tell you where it's working, where it isn't, and the three things to fix first.

3working days
15ppPDF report
30mindebrief call

What this is for

You've set something up - maybe a Claude Project, a Custom GPT, a Jasper account, a couple of half-built prompts in your notes app. It produces output. The output isn't great, and you can't tell whether the problem is the tool, the setup, or you.

The Audit answers that question. Three working days, a 15-page PDF report, and a 30-minute debrief call. You'll know exactly what's working, what's missing, and what to fix first.

What we audit

15 dimensions across the five layers that determine whether AI produces usable output. Each gets a verdict - Good, Partial, or Missing - plus the specific fix needed.

The 15 dimensions
Brand voice captureIs your actual voice documented in a way AI can reference?
Voice rules and word listsSpecific instructions on what to use and what to avoid.
Sample copy as reference materialReal examples AI can pattern-match against.
Business profile completenessPositioning, products, methodology - documented properly?
Pricing and offer clarityCan AI reference your actual products and prices, or does it guess?
Audience persona depthReal personas with language, objections, triggers - or template stubs?
Audience-specific language patternsWords your audience uses, words that repel them.
Competitor awarenessDoes the system know who you compete with and how?
Positioning differentiationCan AI articulate what makes you different - specifically?
Source material structureYour existing content - usable for AI reference, or sitting in folders?
Content tagging and indexingCan AI find the right reference material when it needs it?
Prompt library coverageReusable prompts for your common tasks, or starting from scratch each time?
Prompt quality and specificityAre the prompts working with you, or against you?
Workspace configurationIs the system set up to find and use everything you've built?
Workflow integrationDoes the AI fit your real working process, or is it a separate task?

Each dimension gets a score, a verdict, and a specific recommendation. The three highest-impact fixes get pulled out as the priority action plan.

How it works

Three working days from kickoff to debrief. You don't need to prepare anything beyond access - we work with what's there.

Day 1 - Access and inventory

2 hours of your time

15-minute kickoff call. You share access (Claude Project, Custom GPT, prompts doc - whatever exists). We inventory what's there and run the assessment against each dimension.

Days 2-3 - Assessment and report

Behind the scenes

Each dimension assessed, scored, and written up. Recommendations drafted. Three highest-impact fixes identified. Report built with concrete next steps for each finding.

Debrief - 30-minute call

Live walk-through

We walk through the report together. You ask questions. We give specific, honest recommendations - including whether you need the full Blueprint or whether DIY fixes will do.

Who this is for

  • You've set up AI for your business but the output isn't great
  • You're not sure whether to invest in something bigger or just fix what's there
  • You want an honest second opinion before committing to a £1,497 Blueprint engagement
  • Your team is using AI inconsistently and you need to know what to standardise

Who this isn't for

  • You haven't set up any AI yet - start with the Blueprint instead
  • You want someone to build the fix for you - the Audit identifies, doesn't repair
  • You need certified compliance reporting - this is a practical review, not a formal audit
The deliverable

A real report. With real findings.

Not a PDF template with your name dropped in. Every Audit is run from scratch against your actual setup - the findings are specific, the recommendations are yours to act on. If a dimension's fine, we say so. We don't manufacture problems.

15dimensions
3priority fixes
15ppPDF report

This is usually the point people book a call.

Book a 15-min call →

15 minutes. Honest answer, even if it's “not yet” or “not us.”

Common questions

Things worth knowing before you book.

If something isn't covered, ask on the call. Honest answers, no sales pitch.

I haven't set up any AI yet - should I still get an Audit?

No. The Audit is for businesses that have already tried something. If you're starting from scratch, skip the Audit and go straight to the Blueprint. We'll tell you that honestly on the call rather than selling you something that won't help.

What if my setup is fine and I don't need anything?

Then the report says exactly that. We don't manufacture problems. If most dimensions come back Good or Partial, the report will tell you the few small things worth refining and that's it. You'll know your setup is solid.

It happens. Some businesses have done genuinely good DIY setups. We'll be straight about it.

How is this different from the Blueprint?

The Audit assesses what you've built. The Blueprint builds it for you. The Audit ends with recommendations. The Blueprint ends with a working system.

Some businesses need an Audit to know if they need a Blueprint. Some businesses already know they need a Blueprint. Some businesses just need an Audit and can take the fixes from there.

Does the Audit work with any AI tool, or just Claude?

Any modern context-aware tool. We've audited Claude Projects, ChatGPT Custom GPTs, Jasper workspaces, Gemini Gems, and various brand-voice SaaS setups. The 15 dimensions don't change - what changes is where the gaps tend to be.

If you're using a tool we haven't seen, tell us on the call and we'll confirm whether the methodology applies.

What happens to the £200 credit?

If you take a Blueprint engagement within 60 days of the Audit, £200 comes off the Blueprint price. The Audit pays £197, the Blueprint costs £1,297 instead of £1,497. After 60 days, the credit lapses.

The credit isn't a gimmick to push you into the Blueprint. If the Audit shows you don't need one, we'll say so.

What access do I need to give you?

Read access to whatever AI setup you're using. For Claude Projects, that's a share link. For Custom GPTs, the GPT itself and its instructions. For other tools, the equivalent - whatever lets us see what's been built and run prompts against it.

We sign NDAs as standard. Access is revoked after the Audit ends.

Find out what's actually wrong.

Three working days, £397, honest verdict. If your setup turns out fine, we'll say so - and you'll have it in writing.