A friend who runs leadership development at a mid-sized company called me last month. She was tired.
“We just spent forty thousand dollars on a leadership program,” she said. “Big external facilitator, polished slides, the whole thing. The feedback scores were great. Three months later, my managers are doing exactly what they were doing before.”
I’ve heard a version of that conversation about a hundred times in the last fifteen years. The price tag changes, the facilitator changes, the slides change. The ending doesn’t.
The pattern is so consistent that the McKinsey research on it is almost boring at this point: most leadership programs show no sustained impact 12 months out unless peer accountability is built in. ATD and Bersin both put the share of behavior change attributable to formal training at less than 10%. Everyone in L&D knows this. Everyone keeps buying the same kind of program anyway.
This post is about a different way to do leadership development — peer learning groups. Not as a buzzword, not as a rebrand of something you already do, but as a specific practice with a specific shape. I’ll explain what it actually is, why it works when other approaches don’t, and what it looks like in practice. By the end you’ll have enough to decide whether it’s a fit for your organization.
What is peer learning, exactly?
Peer learning is when a small group of colleagues — usually four to six people — meet regularly to learn with and from each other, guided by structured material on a specific topic.
That’s it. There’s a lot of confusion about the term, so let me say what it’s not:
- It’s not unstructured “talking it out.” There’s a guide, a topic, prompts. The group isn’t just venting.
- It’s not a study group reading a book together. The content is a starting point for application, not the destination.
- It’s not a coaching circle. No designated coach is doing most of the talking.
- It’s not a training program with peer activities sprinkled in. The peer dialogue is the program.
Here’s what it usually looks like in practice. Five managers meet for ninety minutes, twice a month, over four months. They have a discussion guide on a specific topic — say, giving feedback that actually changes behaviour. They open with a check-in: what’s been happening at work since they last met? Then they read six to eight pages of expert content together, and discuss it using the prompts in the guide. They surface real situations — “I’ve got a direct report who keeps missing deadlines and I don’t know how to bring it up without it becoming a thing” — and the group works through them. They commit to one specific experiment to try before the next session. Then they do it again, on a different topic, in two weeks.
That’s the whole methodology. It looks deceptively simple. The reason it works has nothing to do with the elegance of the format and everything to do with what’s happening underneath it.
Why does this work when other approaches don’t?
Three things.
First, the content is applied to actual work. When a manager reads about how to give feedback in a course, the ideas are abstract. When she reads about it sitting next to four colleagues who are about to ask “okay, who has a feedback situation right now?” — the ideas have nowhere to go except onto her actual problems. Application isn’t a separate phase you hope happens later. It’s baked into the session.
Second, the peers create accountability the facilitator can’t. When you commit to “I’ll have that conversation with Marc by next Friday” in front of four people you’ll see again in two weeks, you’re going to do it. When you make the same commitment to a facilitator you’ll never see again, you’re going to forget by Monday. Not because you’re a bad person. Because that’s how human attention works.
Third, the expertise is collective. Most management problems aren’t novel. The four people in the room have probably collectively faced your situation thirty times. They have moves you’ve never thought of. The discussion guide gives them frameworks to make those moves explicit instead of leaving them as folklore.
I’ll add a fourth one that’s harder to articulate: peer learning is the format that respects the actual expertise of the people in the room. Most leadership training treats managers as students who need to be taught. Peer learning treats them as practitioners who need a structured space to think more clearly about what they’re already doing. The difference shows up in how people show up. Less performing, more wrestling.
This isn’t a new insight, by the way. Henry Mintzberg has been writing about it for thirty years. (Henry’s a colleague of mine — we co-created CoachingOurselves together starting in 2007.) The core idea: real managerial learning happens through reflection on experience, in conversation with peers, scaffolded by good ideas. Not through transmission from an expert.
The four ingredients of an effective peer learning group
If you’re going to design or evaluate a peer learning program, these are the four things to look for.
1. The right size
Four to six people. Three is too small — one absence and you’re a pair. Seven and up means someone is always quiet, and people stop bringing the harder stuff. Six is the magic number for getting everyone’s voice in a 90-minute session.
2. A real discussion guide
The single biggest factor in whether peer learning groups work is the quality of the material they’re working with. A bad guide is a bad guide; ten brilliant peers can’t compensate. A good guide does three things: it presents substantive content (not just questions), it provides prompts that move the group from concept to application, and it includes a specific commitment step at the end.
What I see most often when peer learning fails: someone tries to run the format with handouts that are essentially just questions. The conversation goes in circles because there’s no shared substance to react to.
3. Cadence and continuity
You need at least four sessions over a few months for anything to stick. One-offs don’t count. The point is the iteration: try something at work, come back, share what happened, try something else. A program of six to ten sessions over four to six months is the sweet spot for most organizations.
4. Peer co-facilitation, not expert facilitation
This is the one most organizations get wrong. They assume they need a trained facilitator in every group. They don’t. What they need is a light internal facilitator (often rotating among the peers themselves) who keeps the group moving through the guide. The facilitator doesn’t teach. The facilitator doesn’t have answers. The facilitator holds space.
This sounds like a weakness — “wait, no expert?” — and it’s actually the strength. The moment an expert is in the room, the group’s expertise stops being relevant. Peer learning works because the expertise is in the room.
Where peer learning works best
Not everywhere. Some honest scope:
It works really well for:
- Decentralized or global organizations where you can’t realistically send a facilitator everywhere
- Matrixed teams that need shared mental models across functions
- Mid-career to senior managers (people with enough experience to bring meaningful situations)
- Organizations going through change — strategy shifts, restructures, culture work
- Building leadership pipelines without an army of certified facilitators
It’s not the right tool for:
- Technical skill training (use a course)
- Onboarding very junior people who don’t yet have enough experience to draw on
- One-off compliance-style content where there’s a single right answer
- Teams smaller than about four people
- Situations where leadership wants to control the outcome of the conversation
If your situation is in the second list, don’t force peer learning. It won’t work, and it’ll burn the credibility for when you actually do need it.
A real example
Saša Štebe — at the time running Amrop Adria’s practice in the Adriatic region; today she runs her own leadership advisory firm, Leadership Solutions, in Slovenia — started experimenting with cross-company peer learning groups for HR leaders during COVID. By 2021 she’d built a structured program: 220 leaders from about 20 different companies, four sessions over four months, mixed cohorts.
What surprised her wasn’t that it worked. It was the friendship. She told me: “You meet for the first time and get this discussion guide. You just follow it and by the end of it, everyone is best friends. It’s rather amazing.” Leaders found out they were all wrestling with the same things — across industries, across countries, across roles. The shared experience was a bigger deal than any specific framework they discussed.
Six months after the program ended, the participants kept asking to continue. Voluntarily. Without a budget line item. That’s the reliable signal that peer learning is working — when people don’t want to stop.
What this looks like at CoachingOurselves
We make discussion guides — short, well-designed, research-grounded — for over 100 leadership and management topics. Organizations license them, set up cohorts of four to six managers, and run their own programs. No external facilitator required. We support train-the-trainer for the people running the cohorts internally, but the sessions themselves run from the guide.
The topics range from foundational (giving feedback, building trust, managing time and energy) to pointed (leading hybrid teams, the Five Mindsets of a Manager, climate-conscious leadership). Browse the full topic catalogue here — about 100 guides, organized by management challenge.
If this sounds like something that might fit your context, the cheapest way to find out is to come to a session. We run free 60-minute discovery sessions where you actually participate in a peer learning experience — not a sales pitch, an actual session — and then we discuss whether it’s a fit. Book a session here. We’ve been running these for years; the conversion rate from “skeptical L&D person attending a demo” to “this is what we needed” is uncomfortably high.
What to take away
Peer learning isn’t a magic bullet, but it’s a specific practice that solves a specific problem — the gap between training content and actual behaviour change. If you’re tired of running programs that look great on the feedback survey and don’t change anything in the work itself, this is worth considering.
Three things to remember:
- Small groups, structured material, regular cadence, peer-led. Lose any of those four and you’re running a different format.
- The peers are the expertise. Build the program around getting them to talk; the content is just scaffolding.
- Don’t force it where it doesn’t fit. Not every learning need is a peer-learning need.
If you want to keep going on this, the most useful next step is to read about the reflective mindset — the core of how peer learning develops capacity, not just competence — or leading with humility, which is itself one of the most common topics organizations start with.
And if you’re designing a leadership program right now, come to a session. It’s the fastest way to know whether this is the right tool for your situation.
— Phil