1 AI Champion per 15-20 People — Why This Ratio Works

When a founder of a 75-person logistics firm asked me why their AI rollout stalled, the answer became clear: they had 3 champions covering 50 people. The math never worked.

The Cognitive Load of Being an AI Champion

Champions handle 3 core tasks:

1. Use case discovery (30% time)
2. Shoulder-to-shoulder coaching (50% time)
3. Progress reporting (20% time)

At >20 people per champion:

• Coaching quality drops below 10 min/week per employee
• Reported use cases become superficial ("I used ChatGPT to write an email")
• Shadow AI use spikes as employees revert to unsanctioned tools

Manager Scan (What AI Champions Report After Week 1)

From actual 30-100 person company deployments:

• 3-5 concrete workflow automations identified
• 12-15 employees actively experimenting (vs 2-3 in tool-first approaches)
• Typical time savings range: 1.5-3 hours per champion per week

Tool tip (Course for Business): Our 6-week program trains champions to spot automation opportunities using the shoulder-to-shoulder method – no coding required. Map your team's first week →

Why 1:15-20 Matches Human Learning Patterns

1. The Ebbinghaus effect: Spaced repetition requires follow-ups every 5-7 days
2. Dunbar's number: Humans maintain ~15 stable peer relationships
3. Apprenticeship model: Skills transfer best through observed practice (2-3 demonstrations)

Micro-Case (What Changes After 7-14 Days)

A 60-person professional services firm deployed 4 champions (1:15). Within two weeks:

• Finance team automated invoice matching (saving 6h/week)
• HR built a screening workflow that cut interview prep time by 50%
• Leadership received a consolidated view of AI adoption across departments

Note on this case: This example is illustrative — based on typical patterns we observe with companies of 30–500 employees, not a single named client. Specific numbers are rounded approximations of common ranges, not guarantees.

FAQ

Q: Can champions be part-time? A: Yes, but allocate at least 5h/week for coaching. Full-time champions often lose touch with frontline work.

Q: How to pick champions? A: Look for:

• Process-oriented thinkers
• Natural teachers (not necessarily the most technical)
• Employees others already seek for help

Q: What if we can't hit 1:15? A: Prioritize departments with repetitive tasks (finance, HR, support) first. Scale as adoption grows.

Q: How long before seeing ROI? A: First measurable gains appear in 2-3 weeks. Full workflow automation takes 6-8 weeks.

Next Steps

This ratio prevents the two most common AI rollout failures: under-supported adoption and IT-led tool mandates. Start by identifying your natural process improvers.

If you want every employee to ship their first AI automation in five days — book a 30-min call and we'll map your team's first week. Start here →