Training
A ChatGPT course for business, built on your team's real work
Your team very likely uses ChatGPT already. This ChatGPT course teaches them to use it with a method: prompts with context, projects per client or topic, clear rules on data and a check on every answer before it leaves the company.
A ChatGPT course for a team that already uses it
"But we all use it already." It is the obvious objection, and it is usually half true. In many teams, one or two people use ChatGPT every day to good effect, several tried it once with a vague question and gave up, and someone pastes client emails into a personal account without knowing what happens to that data.
Gallup found in May 2026 that 30% of US employees use AI at work several times a week or more, and 15% daily. Use is growing, but unevenly. A study by Harvard Business School and BCG with 758 consultants found that people using AI produced over 40% higher quality work and finished 25% faster, and that outside its strengths AI made results worse.
Our ChatGPT training for companies starts from there. The aim is for each person to know which tasks it helps with, how to ask, what data they may enter and how to check the answer. The course works on your company's documents and cases, with generic examples only to explain a principle.
Three questions for your team before the course
Send them anonymously, through a simple form for instance. In ten minutes you get a picture of real use, and the course is designed from the answers.
Which account do you use ChatGPT with?
Free personal, paid personal or a company account. If many personal accounts show up, the first module will be about data and accounts.
What did you paste into it last week?
Client emails, contracts, staff data, internal figures or nothing sensitive. The answers show which rules still need writing.
Which task did it help most with, and where did it fail?
Tasks that work become examples for the rest of the team. Those that fail usually point to a prompt without context or a task that should not be delegated.
What the ChatGPT course covers
Modules are adjusted to each team. ChatGPT's features change often, so the course focuses on stable uses and we review the content for every edition.
Prompts with context
Audience, goal, tone, reference material and the format of the answer. Practised on your own tasks, such as a sales proposal or a reply to a complaint.
Projects
A project groups chats, files and instructions for one client, case or topic. The team learns to set them up so they stop repeating the context every time.
Files and spreadsheets
Summarising a long report, comparing two versions of a contract, analysing a sales export. And checking every figure before using it.
Search and sources
When to ask it to search the web, how to read the sources it cites and why it is worth opening them before reusing a fact.
Rules on data and accounts
What information can go into which type of account, how to anonymise a case and what never goes in. The course produces a one-page usage policy.
GPTs, plugins and connected apps
What to do with your company's custom GPTs now their retirement has been announced, and when it makes sense to connect ChatGPT to other work tools.
What data can go into ChatGPT, account by account
OpenAI separates personal plans (Free, Go, Plus and Pro) from plans for organisations (Business and Enterprise). According to its help centre, as of November 2026, the organisation plans do not use data to train models by default. On personal plans, conversations may be used for training unless the person switches it off in their data controls.
For a company operating in the EU, that has a practical consequence. If your team handles personal data about clients or staff in ChatGPT, it should do so in a company account with the safeguards the GDPR requires, and not in personal accounts the company does not control. Even then, some data is best kept out, such as health information or documents a contract obliges you to keep confidential.
The course does not replace choosing a plan, which depends on your size and admin needs. It does leave each person with a clear rule: what they can paste, into which account, and how to anonymise a case when in doubt.
Custom GPTs are moving to plugins
Many companies built custom GPTs for repeated tasks: a proposal assistant loaded with their price list, another that writes in the brand's tone. In September 2026 OpenAI announced it will retire them and offer a migration to plugins, which bundle instructions, reference files and connections to other tools. At the time of writing, November 2026, retirement is planned for December 2026 on most accounts, with dates that can vary by workspace.
If your company has GPTs in use, the course includes a review: which still earn their place, which can be replaced by a project with good instructions and which are worth migrating. Always check the notice OpenAI shows in your own account, as that is what sets your date.
How the course runs
It is delivered on our dedicated video platform or in person at your company.
- 01
First meeting
In a free 30-minute video call we look at the accounts in use, the teams involved and the tasks that take most of their time.
- 02
Survey and task selection
We send the three questions and pick two or three tasks per role with you, such as sales follow-ups, client replies or monthly reports.
- 03
Hands-on sessions
Each participant works on their own cases, anonymised where needed. They learn to give context, check the answer and recognise when ChatGPT is the wrong tool.
- 04
Prompt library
Prompts that work are saved in shared projects or documents, alongside the usage policy.
- 05
Follow-up
A few weeks later we run the three questions again and time the chosen tasks. If a practice has not stuck, we reinforce it.
Who it is for and what the price depends on
The course is designed for companies whose teams already use ChatGPT or are about to adopt a plan for the organisation: professional firms, estate agencies, sales teams, online shops, back-office departments. If your company also uses Claude, Gemini or Microsoft Copilot, we can cover them in the same programme.
The price depends on the number of participants, the roles to cover, the format and the length. We send a quote after the first meeting. If you send your team the three questions beforehand, the proposal will start from their answers.
Frequently asked questions
How long is a ChatGPT course for a company?
The basics take a few hours of guided practice. Turning them into habits takes longer, which is why we combine hands-on sessions with a follow-up a few weeks later. The exact length is set by the number of roles and the tasks chosen, after the first meeting.
How much does ChatGPT training cost?
It depends on the number of participants, the roles, the format (in person or video) and the length of the programme. We do not publish fixed rates because each course is designed around the team's tasks. We send a quote after a free 30-minute video call.
Is an online or in-person ChatGPT course better?
Video works well if your team is spread across offices or works remotely, and it allows shorter, more frequent sessions. In person helps get a group started that does not use the tool yet and answers questions as they come up. The two can also be combined.
Do we need ChatGPT Business to take the course?
Not necessarily: the course adapts to the accounts your team uses. But if staff will work with client or employee data, a company account is the sensible choice for the data protection safeguards. In the first meeting we look at the accounts in place and what to review before starting.
What happens to the custom GPTs we have built?
OpenAI announced in September 2026 that it is retiring them, with a migration to plugins planned for December 2026 on most accounts. In the course we review each GPT: whether a project can replace it, whether to migrate it or whether nobody uses it any more. Check the date shown in your own account.
How is this different from a free ChatGPT course online?
An open course explains features with generic examples, which is fine for getting started. Ours works on your company's documents and tasks, leaves written rules on data and a library of tested prompts, and then measures whether the team actually uses them.