Leadership development programs are remarkably effective at helping leaders see themselves more clearly. The problem is that seeing a behavior and changing it are two different things.
Many leaders leave a program with genuine insight into what is holding them back. They understand the changes their new role requires and commit to acting differently. Yet when they return to work, the pressures and habits that shaped their success reassert themselves with surprising speed.
Let’s take the example of a leader we will call Elena. She had been identified as one of the strongest high-potential executives in her company, a global specialty chemicals manufacturer. Elena had spent 15 years there, moved through commercial and operational roles in three regions, and had recently been appointed to lead one of the company’s business units. Her sponsors saw in her the makings of an enterprise leader.
They also saw, more clearly than she did, that what had carried her to this point would not carry her further.
Several months into the new role, she received feedback that surfaced a pattern. She was over-managing her most experienced direct reports, introducing too many ideas in senior team meetings, leaving her team feeling obligated to act on every one of them, and working at a level of operational detail she could no longer afford.
Her HR partner urged her into a leadership development program designed for exactly this kind of transition.
The program did everything a good program does. Three days off-site, skilled faculty, a 360 from her team, structured reflection, peer conversations with leaders facing the same shift. She came away with real insight. She saw the pattern, understood where it came from, and recognized the cost to both her and her team. She made the standard commitments: operate more strategically, delegate more, and stop jumping into operational detail. She returned to her role determined to lead differently.
Within weeks, in the situations that mattered most, she was doing what she had always done.
Patterns of thinking and acting that have worked for decades are not overwritten by a weekend of insight, no matter how good the weekend is. And many of these patterns are reinforced daily by what the leader’s organization rewards and expects, which means the leader is not the only force keeping them in place.
This is the part of the story that leadership development rarely addresses honestly. Elena did not fail because she lacked motivation, and she did not forget her commitments. Her insight was real, and her resolve was sincere. She failed because the situations that triggered her usual responses moved faster than her new awareness could intervene. The VP of sales walked into her office with a problem. Elena had an answer before she had a thought. By the time she remembered her commitment to give her VP more autonomy, she had already given her the answer.
And the speed is only part of the story. The behaviors the program asked Elena to change were some of the ones that built her career: solving problems, generating ideas, staying close to the operational detail. In such situations, letting them go may not feel like growth, but like letting go of familiar sources of competence. The result can be an unconscious resistance to embracing needed changes.
We have seen this pattern in hundreds of senior leaders. It is not a failure of the program. It is a failure of the method the program implicitly uses. Reflection produces understanding. Commitment produces intention. Neither, on its own, produces sustained behavioral change, and the research on how habits actually change tells us why.
Patterns of thinking and acting that have worked for decades are not overwritten by a weekend of insight, no matter how good the weekend is. And many of these patterns are reinforced daily by what the leader’s organization rewards and expects, which means the leader is not the only force keeping them in place.
The most useful way to think about behavior change is as a sequence of stages, not a single moment of resolve. At the first stage, you do not yet recognize the pattern. At the second, you see it only in retrospect. At the third, you begin to anticipate the situations in which it emerges. At the fourth, you can interrupt it in real time. At the fifth, the new behavior has become the default.
A good program will move you to stage three. The remaining work – the work that actually changes how you lead – happens in your real life, in real situations, after the program is over. Almost no program tells you that, or equips you to do it.
When faced with this speed of change, organizations typically try one of two approaches, both of which predictably fail.
The first approach revolves around central planning. Firms will perhaps appoint a Chief AI Officer to design the perfect AI strategy across the company. The problem with this is that this person would need to simultaneously understand AI applications in marketing (monthly changes), legal (quarterly shifts), finance (different vendor ecosystem), IT (different risk profiles), and operations (annual updates). Given how fast things move and how different each domain is, this is at best a herculean task, and at worst an impossible one. As a result, the central planning team becomes a bottleneck rather than a value-add.
The second approach is more a case of chaotic adoption. In this scenario, every team tends to do whatever it wants, with the (again, predictable) lack of coordination that follows. An accounting team solves a problem that finance still struggles with, but they never talk. The company pays for five different AI tools that do similar things. Nothing works together. No one learns from anyone else.
A micro-experiment is not a goal in disguise.
What closes the gap between insight and behavior is something simpler and smaller than another commitment or a more sophisticated framework. It is what we call a micro-experiment: a single, specific, observable act of trying something different in one situation, designed to generate evidence about what actually happens.
A micro-experiment is not a goal in disguise. “I will be more present in my meetings” is a goal. It has no situation, no specific behavior, and no way to know whether you did it. A micro-experiment for the same underlying concern would sound like this: “In my next meeting with my VP of sales, I will ask one question I would normally answer myself, and notice what I want to say while she is talking and what happens in the conversation after I do not say it.”
That sentence does work; the goal cannot do so. It commits you to a specific behavior in a specific situation. It tells you what to attend to, externally and internally. It is small enough that you can risk it, even on a difficult day. And it produces evidence you did not have before: about what triggers your behavior, about what happens when you interrupt it, about what your team does with the space when you leave it open, and about what your organization actually rewards when you behave differently.
Develop key leadership capabilities and accelerate your impact as a leader
Four criteria separate a useful experiment from another resolution. The experiment must be specific enough that you would know whether you did it. It must be connected to a pattern you have actually identified in yourself, not a virtue you would like to acquire.
It must be observable, which means you can report afterward what happened, including what happened inside you, before and after. And it must be low-stakes enough that you can run it without putting the work at risk. If the experiment does not produce the response you expected, that is data. If it does, that is evidence to build on.
Designed this way, an experiment shifts the leader’s stance from discipline to curiosity. The goal is not immediate self-correction, but learning. Where a commitment asks the leader to muster willpower at the moment the old pattern fires, an experiment asks the leader to become curious: to investigate the situation, not to correct herself. Elena’s commitment to “delegate more” required her to remember her intention at the exact moment her old pattern was firing.
The experiment requires nothing of the kind. It is a single planned act, in a single foreseeable situation, with attention to what happens. The energy it requires is small. The energy it returns, in the form of evidence and small wins, is large.
In Elena’s case, three experiments built directly from the feedback she had received would have given her more in three weeks than the program had given her in three days. One on what she does when her VP of sales brings her a problem. One on how she introduces ideas in senior team meetings, and what happens when she names an observation without asking the group to act on it. One on the operational reports she reads and edits, and what happens when she sends them back unmarked. Each one specific, each one low-stakes, each one a source of evidence she could carry into the next experiment.
There is a final, easily overlooked benefit. The experiments work even when they fail. An experiment that produces an unexpected reaction tells you something about your team you did not know. An experiment that triggers your old pattern earlier than you expected tells you something about the situation you had not seen. There is no version of a well-designed experiment that returns nothing. The same cannot be said of a commitment.
The implication for how we design leadership development is uncomfortable and worth saying plainly.
The implication for how we design leadership development is uncomfortable and worth saying plainly. Programs that end at insight and commitment are leaving the most important work undone.
Participants do not know this when they leave, because the program has done what it promised. The patterns that brought them in remain intact, and within weeks the program’s effect erodes into another set of frameworks they can describe but cannot act on. The investment, which is substantial, returns less than it should.
What good programs should add is a structured experimentation phase, owned by the leader and supported by the organization, in which the insight from the program is translated into a sequence of small experiments run in the leader’s actual life over the following months. This is not an add-on. It is the place where behavior change actually happens, and treating it as optional is the reason so many programs feel valuable in the moment and produce so little durable change. A program that sends a participant home with five well-designed first experiments has done more than a program that sends them home with a development plan, even an excellent one.
Homegrown brands are overtaking once prestigious multinationals in China. Winter Nie, Yunfei Feng, and Marco Gadola offer three pathways for global players to respond....
Audio available
Every team has a “mojo” phase. People are curious, meetings have energy, problems feel solvable, and even conflict can feel productive. Then, slowly, something changes... ...
Strategic leadership insight from Moneyball: stop paying for reputations and buy runs by valuing on-base potential and data-driven decision making....
In a world of hyper-pressure where leaders are being pulled in every direction, the challenge is to perform but also to empower, inspire and protect human energy, creativity, judgement, and courage. The...
Explore first person business intelligence from top minds curated for a global executive audience