Your team does not need another AI hype talk.
They need to know what is real, what applies to you,
and what the law now requires.
Clients and companies I worked for
You know AI matters. Nobody in the building can say what to do next, and it feels like the clock is running. These four sessions turn that worry into a sequence: leaders agree on a direction, the business picks its use cases, your people learn the method, and the organisation keeps the record.
Who this is for. Your staff already paste things into ChatGPT. Nobody has told them what is allowed, nobody owns the AI question, and you cannot tell whether competitors announcing AI projects are ahead of you or bluffing. If that sounds familiar, these sessions are built for you.
How the four fit together. The Briefing gives decision-makers one realistic picture. The Readiness Workshop turns it into a scored use-case list with owners. Practical Training equips the people doing the work. AI Literacy covers responsible use and leaves a documented record. Most companies need two of the four; the guide below picks them.
Your leadership cannot agree on what AI should do here. After this half day, you share one realistic picture, know where the risks sit, and leave with the two or three decisions worth taking this year.
See the briefing › Full dayIdeas are scattered across the business and nothing has priorities. After this day, you hold a scored roadmap of your own use cases, with an owner's name against every priority on it.
See the workshop › Hands-onLicences bought, results all over the place. After this training, your team works with a method: which tool for which job, how to ask, how to check the answer, and what never goes in. Practised on their own work.
See the training › DocumentedThe EU AI Act expects your staff to be AI-literate, and you have nothing on file. After this training, every role knows what it may and may not do with AI, and you hold the documented record.
See the training ›Buying the wrong session is the most common mistake here, and it is entirely avoidable.
Since February 2025 the EU AI Act expects staff AI literacy appropriate to their role. This session is built for that, with a documented curriculum and participation record. Operational training, not legal advice, an audit, or a certification of compliance.
Delivered in-house for a whole team, or one to one for an owner or team lead who wants to lead the shift themselves. In person, remote, or as a webinar.
Request a call. Tell us who uses AI today and what you are trying to achieve; you get a straight recommendation, including the case where the answer is none of the four.
Request a callIn-house or remote · built from your context · no obligation
Half a day that leaves your leadership team with a clear view of what AI can realistically improve here, where the risks sit, and which decisions are worth taking this year. Nothing technical required of anyone in the room.
The problem is rarely a lack of interest. It is that everybody in the room has read something different and none of it agrees.
Not a keynote, not a sales pitch, not technical training. No parameter counts, no product demos, no partner commission at the end of the afternoon. Your leadership does not need to understand these systems internally; it needs a reliable sense of what they can be trusted with, in business language.
It is a working session for the people who decide where the money goes, built so that the discussion afterwards is a better one.
Four parts, with discussion running through all of them. The last one is the reason for the first three.
What these systems genuinely do well, what they do badly, and where the line sits, explained in business terms rather than technical ones.
Practical examples from marketing, e-commerce and operations, chosen for your industry and your size instead of lifted from a generic deck.
Data protection, confidentiality, accuracy, the obligations that already apply to you, and where human oversight has to stay regardless.
Open discussion of what your organisation should do next, what deserves a proper look, and what can safely be ignored for now.
The preparation matters more than the delivery. A briefing assembled from a standard deck will be politely received and forgotten by Friday, which is why the session is built from your context before a single slide is written.
A short intake first: what your business does, where the pressure is, what the leadership team actually argues about. The session is built around exactly those questions, with examples from your sector. You will not sit through anybody else's case studies.
You leave able to tell three things apart: what can be automated with confidence, what needs a person checking, and what will produce a convincing answer that is wrong. That one distinction turns AI from a source of worry into a set of decisions.
Real applications in marketing, e-commerce and operations, with honest costs and returns. Including the ones that did not work, which teach more and are missing from most presentations.
Where legal and reputational exposure genuinely sits, what the EU AI Act already asks, and what a sensible internal position looks like. Enough for good decisions, honest about where you need a lawyer instead.
The most valuable part is usually the argument in the last hour. The session is facilitated so it happens, and ends with something written down.
The same vocabulary and the same expectations, which shortens every subsequent conversation on the subject.
What is worth your attention this year, and what is a genuinely interesting development that has nothing to do with your business.
Data, confidentiality, accuracy, oversight and staff obligations, stated plainly enough to be acted on rather than worried about.
And, just as usefully, where it is not, which saves the cost of investigating something that was never going to pay.
In writing, so the discussion survives contact with the following week.
Once the basics are laid out openly, the senior people in the room can ask the questions they have been avoiding.
Seventeen years of marketing and e-commerce work, and a practice that implements rather than commentates. The useful answers to what does it cost and how long does it take come from having done it.
What is not on offer: a certification, a compliance opinion, or a vendor recommendation dressed up as neutral advice.
A real client example belongs here. Left empty on purpose rather than filled with an invented story.
No, deliberately. Nobody needs model internals to make good decisions. Business language throughout, commercial questions only: what would this improve, what would it cost, what could go wrong, what first.
Yes. A short intake covers what your business does, the pressure, and what has been tried; examples come from your industry and size. A generic version would be cheaper and worth less.
The people who decide where budget goes: management plus the affected department heads. Six to a dozen works best. Small enough that everybody speaks, large enough that decisions stick.
The briefing builds shared understanding and ends with a direction. The Readiness Workshop assumes that understanding and spends a day scoring use cases into a roadmap. Many do the briefing first; going straight to the workshop is fine if your team is aligned.
The session material plus a short written summary: opportunities, risks worth attention, recommended next step. Short on purpose; fifty pages would look impressive and get read less.
Request a call. You get an honest opinion on whether a briefing is the right start or whether you are already past it.
Request a callHalf a day · tailored to your business · nothing being sold in the room
A day with the right people in one room, and a scored, prioritised roadmap at the end of it: which of your AI ideas are useful, which are feasible, and which are worth funding first.
Almost nobody arrives at this workshop short of ideas. What they are short of is a way to choose between them that survives a management meeting.
A full day, with the preparation done beforehand so the day itself is spent on work rather than on introductions.
Context gathered in advance, plus short input from the people attending, so the day starts from your reality instead of from a blank flipchart.
What actually happens in the areas under discussion, including the workarounds. This is where most of the real opportunities are found.
Structured and facilitated, with the people who do the work rather than only the people who manage it. Quantity first, judgement afterwards.
Each candidate assessed against business value, feasibility and dependency, in the room, with the disagreements aired rather than smoothed over.
A sequence, a named owner for each priority, and an agreed next action. Without this last step the day is entertainment.
Every idea gets the same three questions. It sounds obvious, and it is remarkable how rarely all three are asked of the same idea on the same day.
In money or hours, honestly, with the assumptions on the wall where everyone can argue. A use case whose value cannot be described is not ready to be funded.
Whether your data, systems, processes and people can support it as they are today, not in an idealised version of your company that exists in a slide.
Dependencies, approvals, risks and the order they impose. This is the question that reorders most lists, because the highest-value idea is frequently the one that has to wait.
The facilitation is the product. Anybody can run a brainstorm; the structure turns forty ideas into six scored candidates and one agreed start, without the loudest voice deciding.
Operational people are in the room on purpose. Leadership knows the strategy; the people doing the work know where time disappears. A list built from only one of those is reliably wrong.
Usable by somebody else, deliberately. If you never speak to me again, the roadmap still works; it was written for your team, not as a pretext.
Named processes, named benefits and named owners, instead of a general intention to do something about AI this year.
The same three questions asked of every candidate, so choosing between them is a decision rather than a preference.
Data, process, tooling and ownership gaps written down, which turns the recurring objection into a work item with a cost attached.
Both were in the room when the list was scored, which removes an argument that would otherwise have taken months.
Whether the next step is an audit, a build or nothing at all, you can explain the reasoning to a board.
Every priority leaves the day owned by somebody, which is the difference between a roadmap and a souvenir.
Seventeen years, and a practice that builds these things as well as discussing them. In the room that is a ruthless feasibility filter: easier to say an idea will not survive your data when you have watched one fail.
No implementation contract waits at the end. Some priorities belong in-house, some with your agency or IT partner, some are worth nothing. The roadmap says which, before anybody quotes.
Not on offer: guaranteed returns, or a scoring model dressed as objectivity. The scores are structured judgement made in the open, and your team is expected to argue.
A real client example belongs here. Left empty on purpose rather than filled with an invented story.
A mix, on purpose: somebody who can commit budget, the managers of the areas examined, and two or three people who do the work daily. Eight to twelve. Management-only workshops produce a list of things management wishes were true.
Very little, deliberately: a short intake, some context, brief written input on where participants' time goes. If preparing for a workshop becomes a project, something has gone wrong.
No. Business language, no technical knowledge required. Feasibility is assessed during scoring; genuine technical questions get recorded as dependencies, not debated by people who cannot answer them.
No, intentionally. The workshop produces a scored roadmap with owners. What happens next is your decision: your team, your partners, or a scoped implementation. The roadmap is written so somebody else can execute it.
The audit is my independent examination of your marketing and e-commerce, ending in evidence and a roadmap. The workshop is a facilitated day with your people, any part of the business, ending in their conclusions. Marketing-shaped problem: audit. Whole-business question: workshop.
Request a call. Bring the ideas circulating in your business; we work out whether a workshop sorts them or you already know the answer.
Request a callA full day · your people in the room · a roadmap you own
Your team leaves with a working method for AI, practised on examples from their own desks, and clear boundaries on where not to use it. In use the next morning, not filed away.
How it usually goes: licences for the whole department, a week-one burst of enthusiasm. By week six, two people use it constantly, three occasionally, the rest opened it twice and went back to how they always worked. Months later nobody can say whether the subscription is worth renewing.
Nothing went wrong there, except a tool was mistaken for a skill. Access to ChatGPT does not make somebody able to get reliable work from it. The missing piece is method: what these systems are good at, how to ask, how to check, what must never be pasted in, and when to close the tab.
That method is what practical AI training teaches, using the work your people already have on their desks rather than a tidy example invented for a slide.
None of the following is anybody’s fault. It is simply what happens when capable people are handed a powerful tool and no method.
Research, first drafts, summarising, restructuring, variants and translation. The tasks these tools genuinely help with, done properly rather than hopefully.
Prompting taught as a method rather than as a collection of magic phrases, so it transfers to whatever tool you use next year.
A shared habit of checking facts, sources and tone before anything leaves the building, which is what protects your reputation.
What may go into these tools and what may not, stated plainly enough that a new starter can follow it in their first week.
The repeatable jobs in your team turned into templates and prompts your people actually reuse, instead of starting from nothing each time.
Most of the value in a department is not with the two enthusiasts. It is with the people who need a method and permission before they will begin.
Built on your team's actual work. Real examples are collected first: the monthly report, the product descriptions nobody enjoys, the customer replies that all say the same thing. Those become the exercises, so on Monday morning everyone already knows where to apply what they learned.
Generic AI training fails predictably: people watch an impressive demo of somebody else's job and cannot see where it applies to theirs. Your own material removes that gap.
Judgement gets as much time as technique. Knowing when not to use these tools is a professional skill, and a team that has thought about it produces better work.
Roles, tools already in use, current confidence levels and real work examples collected from the people who will attend.
Hands-on and practical, working on your own tasks, with enough structure that quieter participants get as much out of it as confident ones.
What good looks like, what is off-limits, and how work gets checked. Agreed in the room and written down before everybody leaves.
Templates, the reference guide, and a short list of workflows worth building next, so the training does not evaporate by the end of the month.
Seventeen years, and daily production use. Training by somebody who only reads about the tools is enthusiastic about what photographs well and quiet about what wastes your team's afternoon.
The tools discussed are the ones tested publicly on this site, disappointments included. Your team gets the honest version, usually including one thing AI was said to do and cannot.
Not on offer: certification, promised productivity gains, or the idea that everybody should use AI for everything. Some should not, and saying so is part of the job.
A real client example belongs here. Left empty on purpose rather than filled with an invented story.
Usually ChatGPT and Claude, plus whatever you have standardised on. Not chosen yet? Settled at scoping; training people on a tool you are about to drop wastes a day.
Yes, and it is why this works. Real examples are collected beforehand and become the exercises. People do not reliably transfer lessons from somebody else's job to their own.
Yes. Sessions are built for the real mix: a couple of confident users, several cautious, a few who never opened the tool. Nobody left behind, nobody bored; that is exercise structure, not slower delivery.
Paid accounts help; free versions are limited and change often, but a session can run on what you have. If a modest licence change would make a large difference, you will hear it.
It covers responsible use, confidentiality and human oversight. If you are booking against a documented obligation, AI Literacy and EU AI Act Training is built for that, with a curriculum and participation record. This session is capability first.
Request a call. Tell us what your team does all day; we work out whether training is the answer or the process underneath is the problem.
Request a callBuilt from your work · hands-on · on site or remote
Employees need to understand how to use AI responsibly, not merely which buttons to press. This EU AI Act training builds practical AI literacy role by role, explains the risks in language people recognise from their own work, and leaves your organisation with a documented record of what was taught and who attended.
AI literacy Your people already use these tools with your data, and most of them were never told what is safe. AI literacy fixes exactly that: knowing what the tools are reliable for, where they go wrong, what must never be put into them, and when a human checks the result. Working knowledge, not engineering.
The EU AI Act made it explicit: since February 2025, organisations are expected to ensure staff have sufficient AI literacy for their role and context. Sufficient is not a checklist, which is uncomfortable and honest: a marketing assistant and a manager using AI in hiring need very different understanding.
The part most providers omit: the literacy expectation currently carries no penalty of its own, so anyone selling this training on the threat of an imminent fine is selling fear. The real reasons are duller and better: your staff already use these tools, your confidential information already goes into them, and if a problem arises you want to be the organisation that trained its people and wrote it down.
The gap is rarely willingness. It is that AI arrived through the side door, one employee at a time, and no policy ever caught up with it.
Four steps. The last one is the one that distinguishes this from an ordinary training session.
Which groups use AI, in what way, with what data, and what level of understanding each role genuinely requires. Existing policies are read rather than assumed.
Content built per user group. Marketing, sales, service, HR and management face different risks and get different material and different examples.
Live sessions, on site or remote, with role-relevant exercises rather than a lecture. People remember what they practised, not what they were shown.
A written curriculum and an attendance record, handed over for your own file, so what was covered and who received it is a matter of record.
The content is built around the two things that actually protect an organisation: people who understand what they are handling, and a record showing they were taught it.
Role-based means what it says: finance and marketing receive different sessions. A session pitched to satisfy everyone is remembered by no one.
Where the assessment finds a gap training cannot fix, a missing policy, a tool creating avoidable exposure, you hear it plainly. It is not solved by another slide.
Everybody working from the same understanding of what these tools are for and where they stop being trustworthy.
People who know what is permitted use the tools properly. People who do not either avoid them or take risks, and both cost you something.
Not abstract principles. The specific situations your people encounter, and what to do in each of them.
Curriculum and attendance documented, so the question of what your organisation did has a written answer.
The assessment output feeds directly into policy and internal guidance, rather than sitting apart from it.
Including the gaps training cannot close, which is more useful than a certificate implying there are none.
This is operational AI-literacy and responsible-use training. It is not legal advice, a legal compliance audit or a certification that guarantees compliance. Operational AI-literacy and responsible-use training. Not legal advice, a compliance audit, or certification. Where you need your specific legal obligations assessed, that is qualified counsel, and you will be told so.
Seventeen years and daily production use: the right qualification for operational training, the wrong one for a legal opinion. The line between the two is drawn openly on this page.
It leads with substance, not penalties. Your people already use these tools; the useful response is making sure they understand what they are doing with your data and reputation.
What is not on offer: legal advice, a compliance audit, a certificate guaranteeing compliance, or the suggestion that a single afternoon settles your governance obligations permanently.
A real client example belongs here. Left empty on purpose rather than filled with an invented story.
No, and nobody honestly can. Compliance depends on how you use AI, your systems and your policies. This training delivers role-appropriate literacy and documentation of what was covered, which supports your readiness work rather than replacing it.
The people who use AI in their work, and the people who decide how it is used. The engagement starts with an assessment, not a headcount: drafting copy and applying AI in hiring need very different understanding.
Yes, and it is why it works: marketing, sales, service, HR and management get material built on their own risks. One session for everybody satisfies nobody and is hard to defend as role-appropriate.
A written curriculum per group, participant materials, and an attendance record, handed over for your governance file. A record of training delivered, exactly what it claims, not a certificate of compliance, which nobody can issue.
Request a call. We look at who uses AI and with what data; you get a straight view of what training would cover and what it deliberately would not.
Request a callRole-based · documented · no compliance guarantees, from anyone honest
