Share

The problem with hybrid coaching: why putting AI and human coaching side by side isn't enough

By Eliza Solomon
Recherches | 3 min readPublished October 6, 2026

Most hybrid coaching runs two separate experiences in parallel.


Neither one knows what happened in the other. EZRA's connected coaching model is built differently. The AI layer and the human coach share information, so progress compounds between sessions instead of resetting.

AI is already changing how people develop at work, and the productivity case is real. A Harvard Business School and Boston Consulting Group study of 758 consultants found AI drove 12.2% more tasks completed, 25.1% faster work, and 40% higher quality output.

But the same study found the catch. When AI gave a plausible but incorrect steer, strategic decision accuracy dropped by 19 percentage points. Being fast and convincing doesn't make AI infallible. Sometimes the most persuasive answer isn’t the right one.

That's where human coaching still matters. A coach can catch a false assumption AI would miss, and guide someone to a breakthrough AI can't get to alone.

AI and coaches do different things well. Which is exactly why combining the two sounds so compelling.

The problem? Most of what gets called "hybrid" today still falls short.

Our new whitepaper, The Problem with Hybrid Coaching, explores why.

Why most hybrid coaching isn't really hybrid at all

In most organizations, "hybrid" means two separate products running alongside each other: a conversation with an AI tool, and a separate conversation with a human coach. Each might add value on its own. But neither knows what happened in the other.

The person being coached is left to join the dots themselves, with development starting and stopping every time they move between the two.

But what if those gaps didn’t exist?

What changes when AI and human coaching are truly connected

When AI and human coaching are genuinely connected, context carries forward instead of resetting. Someone can work through a challenge with AI in the moment, then take it deeper with their coach, without having to start over or explain how they got there.

And it works the other way too. AI can pick up on what's happening between sessions and bring the coach up to speed before they meet. What happens in the coaching session then gives AI more to work with afterward, shaping the support that comes next.

Each becomes more useful because the other is there.

The difference shows up in behaviour too. A BCG 2025 survey of more than 10,000 workers found that only 51% of frontline employees are regular users of AI tools, a figure that has stagnated. EZRA's own data shows the power of combining human coaching with AI: 78% of people respond immediately when AI is introduced alongside a human coach.

The full whitepaper goes deeper into:

  • Why most "hybrid" coaching is really two products bolted together, not one

  • What a genuinely connected development experience looks like in practice

  • How to measure development as an ongoing journey instead of isolated moments of activity

the problem with hybrid coaching to explore why putting AI and human coaching side by side is no longer enough, and where EZRA believes coaching is headed next.

Share