A fixed-scope assessment that ends with a ranked roadmap, an honest read on your data, a policy your team can follow, and a list of what you are already paying for that you do not need. Yours to keep and act on with or without us.
Fixed scope, fixed fee, and a defined set of deliverables at the end. No open-ended discovery.
WEEK 01
We sit with each function and document how work actually gets done — not how the org chart says it does. Volume, time spent, who touches what, and where the manual effort concentrates. Most of the value of this week is that somebody finally writes it down.
WEEK 02
Every AI tool and subscription already in the building, including the ones bought on a personal card. What is being used, what overlaps, what is being spent, and what your team is quietly doing with consumer tools on company data.
WEEK 03
Every candidate workflow scored on volume, rule clarity, error tolerance, data readiness, and metric strength. The output is a ranked list with an expected impact and a realistic effort estimate against each one.
WEEK 04
A sequenced plan, an acceptable-use policy, a data handling standard, the consolidation list, and a presentation to your leadership team. Including, where it applies, the recommendation to do nothing yet.
Written to be used by your team, not to be impressive in a meeting.
Every candidate initiative scored and sequenced, with expected impact, effort, dependencies, and the metric each one would be judged on. The order matters more than the list.
An honest assessment of whether your data can support what you want to do. This is the section that most often changes the plan, and the one most vendors skip.
A short document your team will actually read: what tools are approved, what data may go into them, what always requires a human, and who to ask. Written in plain language.
What you are paying for, what overlaps, what nobody uses, and what to cancel. This frequently covers a meaningful part of the audit fee on its own.
The findings presented to your leadership team, with the reasoning exposed so people can argue with it. Agreement in the room is worth more than a document nobody opens.
What your team is already doing with AI outside any policy. Nearly always more than leadership expects, and the fastest-moving risk in most businesses right now.
A fixed-fee audit that always concludes you should buy the retainer is not an audit. Some of the outcomes we have delivered are genuinely unwelcome, and we would rather deliver them than be wrong expensively later.
The most common one: your data is not ready. If the customer record is spread across four systems that disagree, no amount of model quality fixes it, and the honest sequence is to consolidate first. That is often a Zoho or systems project, not an AI one.
The second most common: you have one clear automation target and no portfolio. In that case you do not need an AI executive at all — you need someone to build the thing. We will tell you that, point you at our build team, and end the engagement there.
What to expect before you commit to four weeks.
It is a fixed fee, quoted after a short scoping call, and it varies mainly with headcount and how many functions are in scope. We quote a number before any work starts and it does not move.
It is not free, and that is deliberate. A free audit is a sales call with a document attached, and it tends to conclude whatever the seller needs it to conclude.
Roughly two hours from each function lead across the four weeks, plus a couple of hours from a project sponsor. We do the documentation, the scoring and the writing.
The one thing we cannot do without is access to the people who actually do the work. An audit conducted entirely with leadership produces a description of how the company believes it operates, which is usually not the same thing.
No, and a meaningful share of audits end there by design. You own every deliverable and can hand the roadmap to an internal team or another partner.
If you do continue, the audit fee is credited against the first three months of a retainer.
Yes, and for larger organisations that is often the better start — pick the function with the most obvious pain, prove the process works, then widen.
For companies under about 150 people we would usually argue for the whole operation, because the highest-value finding is frequently a workflow that crosses two departments and belongs to neither.
Send it over before you book. If it holds up, we will say so and you should not pay us to repeat it.
What we typically find in strategy decks produced elsewhere is a list of technologies rather than a sequenced set of decisions with metrics attached. If yours names the workflows, the order, the owners and the numbers, you are further along than most and probably need building rather than planning.
Four weeks, fixed fee, five documents you keep. Including, if it applies, the recommendation not to proceed.
