Handing employees access to AI tools does not make a team more productive, and most companies discover that about six weeks after buying the licences. Practical AI training gives your people a working method, real examples from their own desks, and clear boundaries, so the tools improve the work instead of quietly complicating it.
Here is how it usually goes. Licences are bought for the whole department. Week one brings a burst of enthusiasm and a great deal of experimenting. By week six, two people use the tool constantly, three use it occasionally for things they would previously have written themselves, and the rest opened it twice and quietly went back to working the way they always had. A few months later somebody asks whether the subscription is worth renewing, and nobody can answer with anything better than an opinion.
Nothing has gone wrong there, except that a tool was mistaken for a skill. Access to a text editor does not make somebody a writer, and access to ChatGPT does not make somebody able to get reliable work out of it. The missing piece is method: knowing what these systems are genuinely good at, how to ask properly, how to check what comes back, what must never be pasted into them, and when the honest answer is to close the tab and do the job yourself.
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.
The session is built around the work your team actually does. Before it runs, real examples are collected: the report somebody writes every month, the product descriptions nobody enjoys, the customer replies that all say roughly the same thing. Those become the exercises.
Generic AI training fails for a specific reason. People watch a convincing demonstration of somebody else’s job, agree it was impressive, return to their desk and cannot see where it applies to theirs. Using your own material removes that gap entirely.
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 than a team that has simply been told to use AI more.
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 of marketing and e-commerce work, and daily use of these tools in real production rather than in demonstrations. That is the relevant qualification here. Training delivered by somebody who only reads about the tools tends to be enthusiastic about the parts that photograph well and quiet about the parts that waste your team’s afternoon.
The tools discussed are the same ones tested and written up publicly on this site, disappointments included. Your team gets the honest version, which usually includes at least one thing they were told AI could do and it cannot.
What is not on offer: certification, a promise of a specific productivity gain, and the suggestion that everybody in your team should be using AI for everything. Some of them should not, and saying so is part of the job.
A real client example belongs here: the team, what they were struggling with, and what changed in the weeks after the session. Left empty on purpose rather than filled with an invented story.
Usually ChatGPT and Claude, since between them they cover most business use, and whatever else your organisation has already standardised on. If you have not chosen yet, that gets discussed during scoping, because training people on a tool you are about to drop is a poor use of a day.
Yes, and it is the main reason this works. Real examples are collected from participants beforehand and become the exercises. People do not transfer a lesson from somebody else's job to their own reliably, which is precisely why generic AI training is so widely attended and so rarely acted upon.
Yes. Sessions are built for the mix that actually exists in a department: a couple of confident users, several cautious ones and a few who have never opened the tool. Nobody is left behind and nobody is bored, which is mostly a matter of how the exercises are structured rather than of how slowly the material is delivered.
Paid accounts help, because the free versions are more limited and change more often, but a session can be run around what you currently have. What you actually need is settled during scoping, and if a modest licence change would make a large difference to what your team can do, you will hear that.
It touches responsible use, confidentiality and human oversight, which are the practical heart of the matter. However, if the reason you are booking training is a documented obligation, the AI Literacy and EU AI Act Training page is the one built for that, including a curriculum and a participation record for your own compliance file. This session is about capability first.
Request a call. Tell us what your team actually does all day, and we will work out whether training is the right answer or whether the real problem is the process underneath it.
Request a callBuilt from your work · hands-on · on site or remote
