// in this post
Nobody publishes what AI adoption work actually costs, including, until now, us. Part of that is a real reason and part of it is just habit. Here's the honest version, and the one number we can actually fix.
Why there's no single number for the whole thing
A full AI embedding programme depends on things that vary too much between organisations to fit one figure: how many people, how mature your current AI use already is, how much actually needs building into your workflows versus how much is training people on what already exists. Quoting one number for all of that would mean either overcharging the businesses that need less, or underquoting the ones that need considerably more. Neither is honest, and the second one is how consultancies end up padding the invoice halfway through to cover a scope nobody priced properly at the start.
So the full programme, the work of actually embedding AI into how your team works day to day, gets scoped once we know what we're scoping. That's normal practice, not evasion, most serious consultancies in this space work the same way. What's unusual is saying so plainly instead of hiding behind "get in touch for a quote."
The number that is fixed
Discovery, the first phase of the work, is priced and scoped the same way for everyone: £2,500, roughly two weeks, fixed scope.
For that, you get two concrete things. A prioritised plan of which workflows in your business are actually worth building AI into, and the order to tackle them in, based on time sat with the people doing the work, not a generic best-practice list. And the strategy and governance policy for how AI gets used across the organisation: what's in bounds, where sensitive data can and can't go, who owns the decisions. Both are built to be useful even if you decide the rest of the work isn't for you.
That second part matters more than it might look. It's the same territory our piece on AI governance for UK SMEs covers, except built around your actual workflows rather than written as a general framework.
Why fixed-price Discovery, then scoped from there
This structure, a small, fixed-cost first step that produces something real on its own, then a proper quote for the rest based on what that step finds, is the model because it's the only honest one available. Nobody, us included, can tell you what embedding AI into your specific workflows will cost before someone's actually looked at them. What we can fix is the cost of finding out.
What it costs to not do this
Set £2,500 against the alternative. A 1,000-person organisation stuck at the typical 20% Copilot adoption rate is carrying roughly £240,000 a year in licences nobody's using, before counting the productivity that never showed up. Discovery is priced to be the low-risk way to find out whether the rest of the work is worth doing, not a hurdle in front of it. If you want to put a number on what stalled adoption is costing your own organisation before you commit to anything, the ROI calculator will build you a defensible figure from your own training spend.
Where to go from here
The full programme walks through Discovery, Embed and Handover in detail. Get in touch if you'd rather talk through whether Discovery is the right first step before committing to anything.
FAQ
How much does AI training and embedding cost?
It depends on your team size, how mature your current AI use is, and how much needs building versus training, so a single figure for the whole thing would be misleading. The one number that is fixed is Discovery, the first phase: £2,500, roughly two weeks, fixed scope. Everything after Discovery is quoted based on what that phase actually finds.
Why doesn't The Human Co. publish a full price list?
Because the honest answer varies too much to fit one. The work after Discovery, embedding tools into real workflows and training people by role, depends entirely on what's actually needed once someone's looked at your specific work, not a generic package. Quoting a single figure before that would mean either overcharging most clients or underquoting the ones who need more.
What does Discovery actually produce for £2,500?
Two things: a prioritised plan of which workflows are worth building AI into and the order to do it in, and the strategy and governance policy for how AI gets used across the organisation, what is in bounds, who owns what, and how data is handled. It is fixed-scope, fixed-price, and structured to be useful on its own even if you go no further.
Is £2,500 for an AI discovery phase expensive?
Set against the cost of adoption stalling, it's small. A 1,000-person organisation stuck at the typical 20% Copilot adoption rate is carrying roughly £240,000 a year in unused licences alone, before counting the lost productivity. Discovery is priced to be a low-risk way to find out whether the rest of the work is worth doing, not a barrier to finding out.