Companies are investing heavily in developing their senior leaders, with global spending on such training estimated at around US$100 billion a year and predicted to continue growing. What hasn’t grown is confidence in the results. Nearly a decade ago, McKinsey found only 11% of executives strongly agreed that their leadership development interventions achieve and sustain the desired results. Today, organisations still rate the majority of their own programmes as ineffective. 

Michael Beer and colleagues at Harvard aptly described this as the “great training robbery”: the fact that capabilities built in classrooms rarely translate to behavioural change. These programmes are often thought-provoking and inspiring, but the sad reality is that the change they aim to inspire evaporates when the trainee returns to their organisation. 

Despite this being well-documented, organisations have done little to design programmes that tackle this issue. Big budgets are spent on training leaders, while the environment that they return to post-training is typically ignored.

Post-training blind spots

There is a famous saying, often attributed to German-American psychologist Kurt Lewin: never send a changed person back into an unchanged environment. The idea is that they will eventually slide back into old habits. Yet most studies into the effectiveness of leadership development only focus on the period during or immediately after delivery. In my research, I followed 18 leaders for several years after each had completed an intensive experiential development programme to understand what happened when a person carried their fragile new self back into their old world.

What I found was that it was indeed the environment they returned to that determined whether any change stuck. It was not just a question of an organisation’s general culture but that the community formed around the individual during the development programme – including the alumni bodies and professional networks – makes a difference.

A gated community

Communities are often the gateway to leadership development programmes. It could be a partner who was a facilitator, a friend who had been transformed, or a boss who saw it as a rite of passage. They extend the invitation and offer reassurance that the unknown, intensive process on the other side of the gate is worthwhile.

It’s well known that communities continue to influence the individual, even after they've passed back through the gate and returned to their daily life. But my research revealed something that has often been overlooked: influence cuts both ways. A community can carry development forward, but it can also hold it back. 

While development often involves some form of transformation, the world resists change in these individuals by holding them to who they were before the training. The wrong kind of community might resist it compassionately by making the person feel so thoroughly appreciated that becoming someone else stops feeling urgent. But development should generate a productive discomfort, a tension that comes from outgrowing the old self. A community that absorbs this discomfort actually absorbs the energy such growth requires. 

The biggest challenge is identifying when a community that looks supportive is actually preventing growth. This is the dark side that the literature on development has not mapped. No engagement survey can see it because the supportive community and the stalling community both produce the same positive engagement, belonging and satisfaction scores.

Take the example of a participant who had spent a decade in one of the world's most respected leadership development communities. The community placed great importance on reflection and talking instead of implementing. It encouraged another workshop, another gathering, but never any doing. Disillusioned after 10 years, she eventually stopped attending. The gate that had once opened to let her into the community had closed and trapped her in. It was preventing her development, but it happened so gradually it took her a decade to notice.

Difference, distance and demands

There are three key ingredients to ensuring that a community does not stall development. The first is difference. Often the communities in my research were built on sameness: same practice, worldview and roles. The problem is a community of people who share the same worldview can validate a transformation, but it cannot stretch it.  

People are often surprised at how strangers can offer as good a perspective, or an even better one, than a close confidant. This is partly because their difference allows them to offer solutions or invite disclosures that are not influenced by shared experiences or emotional connections. 

A learning community made up of people who are different also offers the benefit of distance. Individuals drawn from different companies, roles or locations don’t have to worry about bumping into each other around the water cooler on a Monday morning. 

Distance makes the disclosure safe, and difference makes the perspective new. But there is an important third ingredient. The community also needs to demand that the individual grows, even if that can cause discomfort. Support without demand simply produces stagnation dressed as engagement – lots of talk of change but no real action. Development needs a community that holds the individual but doesn’t blindly agree with them. 

Designing for development

Across executive education, the default move after a transformative experience is to form peer groups – made up of motivated leaders, genuine goodwill and real warmth. Left as voluntary support structures, most just quietly stop meeting. The learning is not that peer groups fail; it is that support was designed in, and obligation was not. The gate needs to be literally built in. 

In a nine-month fully virtual programme that I ran, the gate became the ritual: it opened when the programme started and closed at the end of the day. Participants stepped through the gate into the developmental space and then back out into their lives, again and again for nine months. Years later, alumni still refer to the gate, many stating they miss what stood behind it: not the content, but the held yet demanding space.

Such a space can be designed as deliberately as any building. Three things matter: 

• Diagnose at the identity level, measure after re-entry. Most sponsors profile participants by competencies when they should be profiling them as people: their developmental readiness, motivations, openness and what meaningful progress would look like for them. Don’t measure development at the closing dinner but in the months after their re-entry to daily life, because that is where the return on investment is decided. 
• Build programmes as communities with a gate that opens both ways. Replace one-off events with modules that alternate training and implementation: follow-on programmes, group coaching, compulsory scheduled reasons to return and engage. Create programme-specific communities with real obligations, not institution-wide alumni lists. The rhythm is the point, with the gate opening on a schedule that nobody has to arrange.
• Design for difference and say why. Build development groups across functions and companies, never along reporting lines, and partner with other organisations if your own pool is too familiar. Explain to participants that sameness comforts, difference develops, and distance makes depth safe. Help them choose their post-programme belonging consciously instead of drifting into the warmest space.

None of this argues against belonging; the evidence for connection is overwhelming. They argue for a more precise ambition. Leadership development does not fail for a lack of insight. It fails at the point of return or re-entry, when a new self meets their old environment. Whether it survives depends on the community waiting there: will it only hold the leader or also make demands of them. 

The property industry builds gated communities on purpose. Leadership development builds them by accident, creating safe spaces that are easy to enter but hard to leave. The key is to design the opposite, a community whose gate opens both ways, holding people when they return but demanding that they make changes when sent back out. The change itself takes years. Whether a leader gets those years to develop themselves, in the right type of community, is the part organisations can actually design.