AI consulting: audit, roadmap and fractional Chief AI Officer
We audit where AI genuinely helps your business, rank the options by return against effort, and either hand you the roadmap or run it. The audit is a fixed $1,000 and is credited against your first project. Fractional Chief AI Officer engagements run $5,000 to $30,000 per month.
What you get
A ranked opportunity list, not an inventory — Every candidate use case scored on annual value, effort, risk and how long before it pays for itself — then ordered. The ranking is the deliverable; a list of forty possibilities is what everyone already has.
A costed roadmap with real prices — Each recommendation carries our published price range for the work: agent build, automation, software or search. You can take it to another vendor and compare like for like, and we would rather you did that than sign something you cannot price.
The honest 'do not do this' list — The things that will not pay for themselves at your scale, and the things where an existing tool you already pay for would solve it if somebody configured it properly. This section usually saves more money than the recommendations make.
A data and systems readiness assessment — What data exists, where it lives, whether it is accurate enough to act on, and what would have to be true before an agent or a pipeline could rely on it. Most failed AI projects are actually failed data projects.
A build-versus-buy call on each item — For every recommendation, an explicit position: buy an existing tool, configure something you already own, or build. With reasons, so you can disagree with the reasoning rather than the conclusion.
A measurement plan defined up front — What number should move, who owns it, what the baseline is today, and by when. Defined before the work starts, because a pilot with no pre-agreed measure always gets declared a success and never gets funded again.
Governance sized to your business — Where a human must approve, what must be logged, what data must never reach a model, and who owns the answer when a customer asks whether they spoke to a machine. Practical rules, not a policy framework.
Fractional ownership if you want it — On a fractional CAIO engagement we hold the roadmap, run vendor selection, chair the reviews and are accountable for the numbers we defined — with our own delivery teams available underneath at published prices.
Published pricing
AI audit (productized) — $1,000 one-time. Interviews, systems and data review, ranked opportunity list, costed roadmap, do-not-do list, measurement plan. Credited in full against your first project. Timeline: 2–3 weeks.
Roadmap execution support — Quoted per project. We deliver items from the roadmap at our published service prices, or manage another vendor doing so. Timeline: Per project.
Fractional CAIO — advisory — — then $5,000–$12,000 / month. Roadmap ownership, vendor selection, architecture review, monthly leadership session, quarterly reset against the measurement plan. Timeline: Month to month.
Fractional CAIO — embedded — — then $12,000–$30,000 / month. The above plus weekly working time inside your team, direct accountability for delivery, and priority access to our build capacity. Timeline: Month to month.
How the work actually runs
01. Interviews (Week 1) — Conversations with leadership and, more importantly, with the people doing the work: what takes the longest, what gets redone, where the queue forms, and which system everybody complains about. Leadership knows the strategy; the front line knows the constraint. Output: An interview record and a first list of candidate constraints, written in the business's own words.
02. Systems and data review (Week 1–2) — What systems exist, what data they hold, how accurate it is, what integrates with what, and where the manual bridges are. We look at the actual data rather than the data dictionary, because those are frequently different documents. Output: A systems map with integration points, data quality notes, and the gaps that would block each candidate use case.
03. Rank and cost (Week 2) — Each opportunity scored on annual value, implementation effort, risk and payback period, then ordered. Every item is priced against our published ranges for agents, automation, software and search, so the roadmap is a budget rather than a wish list. Output: The ranked, costed roadmap plus the do-not-do list and the build-versus-buy call on each item.
04. Present and argue (Week 3) — A working session, not a presentation. We expect the ranking to be challenged, and the ranking that survives an hour of argument with the people who run the business is worth considerably more than the one we arrived with. Output: A final roadmap you have edited, with an agreed first project and a measurement baseline recorded before it starts.
05. Own it, or hand it over (Ongoing or ended) — Either the engagement ends and you execute — the roadmap is priced so another vendor can quote from it — or we continue as fractional CAIO, holding the roadmap, selecting vendors and being accountable for the measures we defined. Output: Either a clean handover, or a monthly cadence with named owners and a quarterly reset.
Who this is for
Leadership teams under pressure to have an AI answer, who want a defensible one rather than a pilot that goes nowhere.
Businesses that have already run two or three AI experiments and cannot tell whether any of them worked.
Companies with a real operational constraint and no view on whether AI, automation or software is the correct tool for it.
Who this is not for
Organisations wanting a slide deck for a board meeting. The audit is a working document with prices in it and it is not designed to be presented.
Teams looking to validate a decision already made. We will tell you if we think it is wrong, which is the whole point of paying for the opinion.
Anyone expecting model research or bespoke training runs. Our leverage is applied delivery, not frontier research.
The receipts behind those claims
58 repositories behind the advice — The advice comes from a portfolio of 58 repositories, including production agents, native mobile apps, automation engines and a shipped Unity 6 title. We recommend work we can also do, at prices we publish.
2,245 commits since 13 June 2026 in one client engine — One client codebase carrying roughly 487,000 lines, 239 production tables and 1,378 static pages, built and operated by the same people writing the roadmap. Strategy from people currently in the code.
$1,000 fixed audit price, credited back — Boutique consultancies commonly bill $150 to $350 per hour and top-tier strategy firms $300 to $900. A fixed thousand dollars, credited in full against your first project, removes the incentive to make the engagement longer.
4 delivery lines the roadmap can price against — Agents, automation, software and AI search, all with published price ranges. A roadmap item with no price attached is an opinion; one with a range is a decision you can make.
Against the market
Boutique consultancy hourly rate — market: $150–$350 / hour; here: $1,000 fixed for the audit, credited back
Typical deliverable — market: A strategy document; here: A costed roadmap priced against work we can also build
What is actually in a $1,000 AI audit?
Interviews with leadership and the front line, a review of the systems and the data actually in them, a ranked list of opportunities scored on value against effort, a costed roadmap priced against published ranges, an explicit do-not-do list, a build-versus-buy call on every item, and a measurement plan with a baseline recorded before any work starts. It takes two to three weeks and it is deliberately fixed-price. An hourly audit rewards length; a fixed audit rewards getting to the ranking. The fee is credited in full against the first project you commission, which means if the audit says the right first move is a $12,000 automation, you pay $11,000 for it. What it is not is a slide deck for a board. It is a working document with prices in it, written so that another vendor could quote from it. That is intentional: a roadmap you cannot take elsewhere is a sales instrument, not a deliverable.
What does a fractional Chief AI Officer actually do?
Holds the roadmap and keeps it current as the business changes. Runs vendor selection and reviews architecture before commitments are made rather than after. Chairs the monthly review where the pre-agreed numbers are checked. Kills projects that are not working, which is the job most organisations have nobody to do. And is accountable for the measures defined in the audit rather than for activity. It exists because the alternative — hiring a full-time AI leader — costs several times more and is difficult to get right when the role's scope changes every six months. Advisory engagements run $5,000 to $12,000 per month; embedded engagements with weekly working time inside your team and direct delivery accountability run $12,000 to $30,000. The distinguishing feature versus a traditional consultancy is that our own build teams sit underneath at published prices. You are not being advised by someone whose only product is more advice.
What do AI consultants charge, and why is this priced differently?
Boutique consultancies typically bill $150 to $350 per hour. Top-tier strategy firms bill $300 to $900. On those models a thorough audit is a five-figure engagement before anybody has built anything, and the deliverable is a document. We price the audit at a fixed $1,000 and credit it against the first project, because our economics come from delivery rather than from the assessment. That alignment is worth stating plainly: it means we have no reason to extend the audit, and it also means you should read our recommendations knowing we would be happy to build them. The defence against that bias is that every roadmap item is priced against a published range, so you can take it to another vendor and compare.
You are a build shop. Your audit will recommend building.
A fair concern and the reason for two structural defences. First, the do-not-do list is a required section and it is usually the longest one, because most candidate use cases do not pay for themselves at SMB scale. Second, every recommendation carries a published price range, so you can take the roadmap to any other vendor and get a comparable quote. If our recommendation only holds while we are the only one pricing it, it was not a recommendation.
$1,000 sounds too cheap to be serious.
It is priced as a qualification step rather than a profit centre — boutique consultancies bill $150 to $350 an hour and top firms $300 to $900, so this is a few hours of either at fixed price. We make our money on delivery, and the audit is credited in full against the first project. If you would prefer to pay more, the fractional engagement starts at $5,000 a month and is where the sustained work actually happens.
How much does AI consulting cost?
TrainYourAgent charges a fixed $1,000 for a productized AI audit, credited in full against your first project. Fractional Chief AI Officer engagements run $5,000 to $12,000 per month for advisory and $12,000 to $30,000 for embedded work. For comparison, boutique consultancies typically bill $150 to $350 per hour and top-tier strategy firms $300 to $900.
What is a fractional Chief AI Officer?
A part-time senior owner of your AI roadmap: holding the strategy, selecting and reviewing vendors, chairing the monthly review against pre-agreed measures, and killing projects that are not working. It costs a fraction of a full-time hire and suits organisations whose scope of AI work changes faster than a permanent role can be defined.
How long does an AI audit take?
Two to three weeks: roughly a week of interviews with leadership and the front line, a week reviewing systems and the data actually inside them, and a final working session where the ranked roadmap gets argued over and edited before it is agreed.
What do you deliver at the end of an AI audit?
A ranked opportunity list scored on value, effort, risk and payback; a costed roadmap priced against published service ranges; an explicit do-not-do list; a build-versus-buy call on each item; a data and systems readiness assessment; and a measurement plan with a baseline recorded before work starts.