How to plan an AI workshop your team can actually use

An AI workshop should leave people able to do something useful. Start with that change, then choose the tools and format around it.

This guide helps you turn “we should do something with AI” into a clear brief: who it is for, what they will practise, and what they should be able to do afterwards.

Start with what people should be able to do

Complete this sentence before writing an agenda:

By the end of the workshop, participants will be able to…

Make the answer observable. “Understand AI” is broad. “Turn meeting notes into a first draft of an action plan, then check it for accuracy” gives everyone a clearer destination.

Other examples might be drafting a reusable research prompt, building a simple internal prototype, or mapping an automation. These are possible workshop outcomes, not promises that every team needs the same session.

Bring one workflow from your actual work

Pick a task people recognise. Describe its starting point, the inputs it needs, and what a useful result looks like. Bring a representative example that can be shared during the session.

• Starting point: what do people receive or collect?
• Current process: what do they do with it today?
• Desired output: what should the finished work contain?
• Quality check: how will they decide whether the output is useful?

The quality check matters. A polished answer is only a starting point; participants should also practise reviewing, questioning, and improving it.

Choose the tools around the task

Start with tools the team already has access to, then identify anything else the exercise needs. Confirm that participants can sign in and use the relevant features before the workshop.

A focused session can use one tool well. A longer training can connect several tools around a workflow. The tools page is a useful place to explore the options, from AI assistants and coding tools to automation and data platforms.

Match the duration to the scope

Leave room for a demonstration, an exercise, discussion, and questions. A list of tools on its own does not tell you how much practice people will get.

In my workshop planner, you can choose from 60 minutes to 20 hours. Training above two hours is split into multiple sessions. Here are some ways to organise that time:

These are planning examples. The final agenda and session schedule should follow your team’s goals, starting point, and availability.

Plan for practice and the next working day

Give participants a concrete task, time to try it, and a way to compare what happened. Useful discussion prompts include:

• What did the tool help with?
• Where did the output need human judgement?
• What would you change before using this in your own work?
• What will you try again after the session?

End with something people can take away: a prompt, a checklist, a prototype, or a workflow they have worked on. Agree on a small next step they can try in their normal work.

Write a brief you can actually send

Use this as a starting point. You do not need to have every answer before a discovery call; even a rough brief makes the conversation more specific.

Team or audience:
Number of participants:
Current experience with AI:
One workflow we want to work on:
What participants should be able to do afterwards:
Tools we use or want to explore:
Preferred total duration:
Preferred session length:
Online or in person:
Possible dates:
Anything else the trainer should know:

Ready to put a format around it? Choose your tools and duration in the workshop planner, or book a free discovery call to work through the brief together.