What you will learn
You have been experimenting with AI as a leadership team. Now you need a structure that scales what you have learned across the organisation. That structure is a Champions group, and this page shows you how to build one.Understand the model
Learn the funnel that moves AI from leaders to the broader team
Define roles
Know the difference between a Driver and a Champion
Size your group
Work out how many Champions your organisation needs
Set up governance
Create guardrails that enable experimentation without risk
The adoption funnel
AI adoption does not happen by licensing a tool and sending a login link. It moves through layers.How most organisations do it
Buy licences. Send an email. Hope people use it. Three months later, check usage reports and wonder why adoption is flat.
How it actually works
Leaders learn first. A small Champions group stress-tests tools for your specific context. Then the broader organisation gets onboarded with real use cases, not generic training.
Drivers vs Champions
These are two distinct roles. Getting the distinction right matters.The Driver
The Driver
The Driver is the person who owns the AI adoption programme. They coordinate the Champions, set the agenda, track progress, and report back to leadership. Think of this as the head Champion.In a small organisation, the Driver is often one person. In larger teams, it might be a pair. The Driver needs to know the business deeply enough to make informed decisions about which tools and use cases are worth pursuing.Key responsibilities: Set the agenda for Champion meetings. Track licensing and tool decisions. Escalate governance decisions to leadership. Maintain the Opportunity Tracker.
The Champions
The Champions
Champions are the hands-on testers. They sit across different teams and functions, experimenting with AI in the context of their actual work. They are not IT specialists. They are curious people from operations, marketing, finance, HR, and other departments who are willing to try new things and share what they find.Key responsibilities: Test use cases in their own workflows. Compare tools against specific criteria. Share wins and learnings at huddles. Flag risks and limitations. Help onboard the broader team when it is time to scale.
For a deeper guide on what makes a great Champion and how to select them, see Identifying A Good Champion in the Accelerator resources.
Sizing your Champions group
A good rule of thumb is around 5% of your organisation. For a 45-person company, that is 4 to 5 people. For a 200-person company, aim for 8 to 10. The group should represent a spread of functions. You want someone from finance seeing different opportunities than someone from operations or marketing. Cross-functional representation is what makes the Champions group valuable.Too small
One or two people carry all the load. They burn out. Adoption stalls when they get busy with their day job.
Right size
Enough people to cover key functions. Peer accountability keeps everyone moving. Workload is shared.
Too large
Meetings become unfocused. Decision-making slows. The group feels more like a committee than a task force.
What Champions actually do
Champions are not just “people who like AI”. They have a specific brief.Test use cases for your industry
Test use cases for your industry
Generic AI demos look impressive, but your business has specific workflows, data types, and compliance needs. Champions stress-test tools against real tasks in your context. A financial services firm needs different things from AI than a construction company.
Compare and evaluate tools
Compare and evaluate tools
Champions should have access to multiple tools so they can compare. One organisation tested Gemini, Claude, ChatGPT, and Copilot against three criteria (research, analysis, and content creation) and found that different tools won for different use cases. That insight only comes from structured comparison.
Track what works
Track what works
Every use case that works goes into the Opportunity Tracker. Every one that fails gets documented too. This creates the evidence base for scaling decisions and budget conversations.
Share and teach
Share and teach
Governance without bureaucracy
Governance is not about slowing things down. It is about making sure your organisation adopts AI safely while still moving fast.What governance covers
Which tools are approved. What data can be used. Who reviews new use cases. How incidents are handled. How tool decisions are made and reversed.
What governance does not mean
A 50-page policy document no one reads. Approval committees for every prompt. Blocking experimentation until legal signs off. Making people afraid to try things.
Start with the AI policy framework from the onboarding module. Your Champions group should review and adapt it as they learn what works in practice. Governance is a living document, not a one-off exercise.
Tool licensing strategy
The AI market moves fast. Lock-in is risky. Keep tools on month-to-month contracts wherever possible. This gives you flexibility to switch if a better option emerges or if a tool underperforms its promises. Annual contracts save money on paper, but they cost more in practice if you are stuck with a tool your team stops using after three months. Give Champions access to two to three tools even if you have a primary one. Comparison is how you make informed scaling decisions.Test your understanding
Test your understanding
Quick checkpoint
Funnel understood
You can explain the Leaders, Champions, Broader Org adoption model
Roles defined
You know the difference between a Driver and a Champion
Group sized
You have estimated how many Champions your organisation needs
Governance framed
You have a starting point for tool governance and licensing
Next: set your operating rhythm
Establish the cadence that keeps your Champions group moving