AI Readiness Audit
You assay the ground before you open the mine.
A fixed-fee assessment that grades your data, team, processes, and tooling, ranks the AI opportunities actually on the table, and hands you one clear place to start, so your first project is the right one.
Assay report
6 samples · graded
Headline grade
61/100
Worth a pilot. Not worth a mine. Fix the processes first.
1–2 wks
Start to report
Fixed fee
No open meter
6
Samples graded
1
Place to start
The brief
The expensive way to learn is to build the wrong thing
The most common AI failure isn't a bad model. It's a good build sitting on a shaky foundation, or the right technology aimed squarely at the wrong problem. Both are only obvious afterward, which is what makes them expensive.
An assay gets you the evidence first: what the ground will carry, what it won't, and where the payoff actually sits. It costs a fraction of a failed pilot, and the thing it usually prevents is that pilot.
You walk out with a grade, a gap list, and a starting point. No deck of maybes.
What the report is for
A number, not a vibe
You leave with a grade on every dimension and the gaps named in order of how much they block you. Nothing in the report needs interpreting by the person who wrote it.
A fixed fee and a fixed edge
A defined engagement with a defined deliverable. There is no meter running and no discovery phase that quietly becomes the project.
It ends in a decision
The report names the one highest-payoff place to start, what to fix before that, and what to leave alone entirely, instead of a menu of options to weigh up later.
The samples
Six cores, taken from six different depths
Data foundation
What data you actually hold, how clean and reachable it is, and whether it can feed a model without a rebuild first.
Team & skills
The capability already in the building, the gaps around it, and what you'd need to hire, train, or borrow.
Processes
Which workflows are ready to be automated or augmented, and which ones need straightening out before AI touches them.
Tooling & infrastructure
Whether your stack, security posture, and integrations can carry an AI feature in production, beyond a demo.
Use-case clarity
The opportunities genuinely on the table, ranked by value against feasibility, including the ones worth skipping.
Risk & governance
Data, privacy, and compliance exposure, so adoption doesn't walk you into something expensive later.
The fortnight
Two weeks, and almost none of them are yours
Interviews
Focused sessions with the people who own the data, the tools, and the workflows in question.
Review
We work through your data, systems, and current processes against the readiness framework.
Grading
Each dimension is graded, the gaps are named, and opportunities are ranked by payoff against effort.
The report
You get the grades, the gap list, and the one place to start, walked through live and then handed over.
The method
How a grade gets arrived at
Interview
We talk to the people who own the data, the tools, and the workflows, as well as the sponsor who commissioned the audit.
Review
Your data, systems, and processes get worked through against a readiness framework instead of a first impression.
Grade
Every dimension gets a number against published bands, the gaps get named, and the opportunities get ranked.
Report
A grade, a gap list, and a prioritized first step, written so your engineers and your board can both act on it.
How we grade
The bands are published before your grades are
You can read these before you commission anything, which is the point. A scale defined after the fact can be bent to fit whatever conclusion suits the person selling the next phase.
This dimension can carry a real AI feature as it stands. Build on it. It isn't perfect, but it isn't what's holding you back.
Workable, with known cost. You can proceed here if you go in aware of the friction, or spend a little to move it up before you lean on it hard.
Building on this is how pilots fail. This one gets fixed first, and the report says exactly what fixing it looks like and what it costs.
What the grade changes
What two weeks buys you
The first project is the right one
Most AI budgets aren't lost to a bad model. They go on a good build sitting on a shaky foundation, or on the right technology aimed at the wrong problem.
The argument gets a number
Two people who disagree about whether you're ready are usually both guessing. A graded dimension ends that conversation with evidence.
You find out cheaply
An audit costs a fraction of a failed pilot, and the thing it most often saves you from is that exact pilot.
The board gets a straight answer
One page that says where you stand and what happens next, in language that doesn't need an engineer in the room to translate it.
Selected work
Grounds we've read before
The report
A document you can act on this week
No sixty slides of theory. A tight, evidence-backed document that says where you stand and what to do first, walked through live and then yours to keep.
- 01An overall grade with the per-dimension breakdown behind it
- 02A named list of gaps, ranked by how much each one blocks you
- 03A shortlist of AI opportunities scored on value against effort
- 04The single highest-payoff place to start
- 05The risk and governance issues to close before you scale
- 06A plain-language summary your leadership can act on directly
Where the samples come from
Every sector is barren somewhere different
Professional services
Where the data is in documents and the question is whether any of it is structured enough to be worth modeling.
Retail & distribution
Rich transaction history, usually strong on data and weaker on the processes that would act on a prediction.
Healthcare & regulated
Where governance is graded as hard as capability, because a use case you can't defend isn't a use case.
Software & platforms
Strong tooling, but use-case clarity is often the barren sample: plenty of ideas, no ranking between them.
When to run it
Before you invest
You know AI matters and you'd rather not learn which project was wrong by funding it first.
After a stalled pilot
Something fizzled and nobody can say whether it was the idea, the data, or the ground it was built on.
Under board pressure
Leadership wants a plan next quarter and you need an evidence-based read to build one on instead of a deck of maybes.
Fixed fee
Find out what the ground will carry before you dig it up.
Two weeks, a fixed price, and an answer that's allowed to be no.
Choosing Flaidex for this
An assayer with nothing riding on the answer
We do not sell the shovels
The audit is a fixed-fee, standalone deliverable. Sometimes the honest recommendation is to fix your data, and sometimes it's to skip AI this year. That answer is worth exactly as much as the other one.
The bands are published
You see what pay, marginal, and barren mean before you see your own grades, so the scoring can't be reverse-engineered to suit a conclusion.
We talk to the people as well as the sponsor
The person who owns the workflow knows things the org chart doesn't. The grade reflects the building as it actually runs.
Small does not mean skip it
Smaller teams have the least budget to waste on the wrong first thing. We scope the audit to your size so it stays proportionate.
Working with us
What the engagement itself is like
Low lift on your side
A handful of short sessions and read-only access. The audit is designed so it never becomes a project that consumes the team it's assessing.
Written for two audiences
One document that an engineer can act on and a board can read, instead of a technical annex nobody opens and a deck that says nothing.
No obligation attached
The report is yours. Take it to us, to your own team, or to somebody else. It's written to be useful either way.
Questions
What people ask about the audit
Q1How long does the audit take?
Typically one to two weeks end to end: a few focused interviews, a review of your data and systems, and then the graded report. It's designed to be fast and low-lift on your side.
Q2What do we actually get at the end?
A grade across each dimension, a ranked list of gaps, a shortlist of AI opportunities scored by value and effort, and a clear recommendation for where to start, written so both your technical team and your leadership can act on it.
Q3Is this just a sales pitch for more work?
No. It's a fixed-fee, standalone deliverable. The recommendation might be to build something, to fix your data first, or to hold off on AI for now. We tell you what the evidence says, never what sells the most follow-on work.
Q4Who needs to be involved on our side?
Usually whoever owns the data, the key tools, and the workflows in question, plus a leadership sponsor. It's a handful of short sessions, never a project that consumes your team.
Q5We're a small team. Is this overkill?
Not at all. Smaller teams benefit most from not wasting their first AI budget on the wrong thing. We scope the audit to your size so it's proportionate and practical.
Q6What if we're clearly not ready?
Then that's the most valuable thing you can learn, and you've learned it before you spend. We'll show you exactly what to shore up, usually the data and one or two processes, so you're ready to move with confidence next quarter.
Elsewhere in AI Transformation & Consulting
Let's talk
Running a large platform, shaping a first MVP, or getting a product ready for a funding round? Tell us where you are. We'll shape the process around it, and stay with you after launch.
