Leadership Labs: Developing Management Potential at AELO Swiss Academy

The most useful moment in any lab I facilitate usually arrives around minute forty of a ninety-minute session. Not during the polished opening exercise, but in the middle of something going slightly wrong. Last spring, a participant named Sofia, a senior data engineer five weeks into her first team lead role, had to tell a long-serving colleague that the reporting workflow he had built would be replaced. She opened with small talk, buried the message under praise, and by the end her counterpart, a peer playing the colleague from a written brief, was genuinely unsure whether anything was changing at all. In the debrief, Sofia said she had not wanted to hurt him. Someone asked her what would happen to him in six months if nobody told him clearly. She sat with that for a long moment, and the real work of the session began.

That sequence, rehearsal, wobble, honest debrief, is the whole idea behind the leadership labs at AELO Swiss Academy. Management potential is not revealed in a seminar and it is not certified by a certificate. It is built the way most complex skills are built: through repetitions at the edge of someone's current ability, followed immediately by feedback from people who were watching closely.

Why a lab, not a course

The word lab is deliberate. Nobody learns to swim from a lecture, and the research on deliberate practice has made the same point about professional skills for decades: improvement comes from focused repetitions with feedback, not from exposure to content. Management has almost no natural practice grounds. A new manager's mistakes are visible to their team and often expensive, so most first-time leads default to caution. Caution reads as either absence or micromanagement, and both erode trust, which makes the next conversation harder, which pushes the manager further into caution. The cycle is quiet and common.

A lab breaks that cycle by creating a middle space. The stakes for learning are high, the stakes for career are zero. In our standard format, ten to twelve participants meet for six sessions over twelve weeks, with two facilitators in the room. Each session is built around simulations drawn from the participants' actual working lives, which matters more than any facilitation technique I know.

The shape of a cycle

The twelve weeks open with a diagnostic: a structured interview with each participant, a short self-assessment, and a conversation that includes the participant and their sponsor together. That last detail is non-negotiable for us. No back-channel assessments, no manager describing the participant's weaknesses in a room the participant never hears about. Candor later depends on transparency now.

Sessions run every two weeks rather than weekly, because the space between labs is where the actual development happens. Between sessions, each participant runs one real conversation at work tied to the lab theme and writes half a page on what happened. Those half-pages become raw material for the next lab, and they are often more revealing than anything we observe in the room.

A single ninety-minute lab tends to follow this rhythm:

A short warm-up scenario, about ten minutes, so people are speaking before they start overthinking. The main simulation, fifteen to twenty minutes, with one participant in the chair, a briefed counterpart, and observers each assigned one narrow thing to watch. The debrief, thirty-five to forty minutes, which is deliberately the longest block. Peer commitments, where each observer names one experiment they will try in their own next real conversation. A transfer assignment, agreed out loud, connecting the session to a live situation at work.

The proportions are the point. Early on, we ran content-heavy sessions with short debriefs and learned quickly that we had it backwards.

Scenarios come from the participants' own floors

Every scenario in a lab is harvested from the diagnostic interviews, anonymized, and rebuilt with specific pressure. A reorganization announcement that the participant did not design and does not fully agree with. A high performer who has gone quiet since being asked to mentor a struggling colleague. Two strong specialists in open disagreement about a technical approach while a deadline holds. A team member who agrees warmly in the meeting and does the opposite afterward. These are not exotic cases; they are Tuesday.

The person playing the counterpart receives a one-page brief that includes a hidden constraint. One brief might say: you will not mention the real deadline unless you are asked directly twice. Another: you are more worried than you are letting on, and you will only say so if the participant stops filling the silence. Hidden constraints teach something slides never can, that information in real rooms is asymmetric and the other person's cooperation is not automatic.

The Swiss context shapes this work in useful ways. Our cohorts typically mix participants from German-speaking and French-speaking regions, sometimes with colleagues managing teams across borders as well. The labs run in English, but many of the real conversations participants face happen in their first language, so wherever possible we rehearse in the language the actual conversation will use. A participant managing people scattered between Basel and Geneva, or across a border into Germany, faces asynchronous and cultural load that a purely local program would never surface, so the scenario bank includes messages that must land in writing and calls that must land across a time gap. There is also a cultural fit worth naming: in a country where structured apprenticeship is an ordinary path to mastery, people tend not to find structured rehearsal of management craft strange. They recognize it.

The debrief is the actual product

Newcomers assume the simulation is the event. The debrief is. We follow a fixed sequence because the sequence protects honesty. The participant speaks first, answering one question: what were you trying to do? Then observers report behavior, not motive. "At two minutes in, you asked a closed question and answered it yourself" is usable feedback. "You are not open to input" is a diagnosis, and diagnoses shut people down. Then the counterpart describes what it felt like to receive the conversation, which is often the moment faces change. Finally, we replay one segment of video, about ninety seconds, chosen with the participant rather than for them.

The facilitator's own voice is the least valuable one in the room. If we offer a verdict first, the group optimizes for agreement, and the debrief dies. Our job is to keep the sequence moving and to ask the second question after everyone has asked their first.

Recording is consent-based, used only inside the debrief, and deleted at the end of the cycle. One or two participants decline in every cohort, and their labs work fine on peer observation alone, so video is an option, not a pillar. What is a pillar is the rule set agreed in session one: feedback is specific, it is timed, and it is offered as an experiment to run next time, never as a grade.

What we measure, and what we refuse to

We track five observable dimensions across the twelve weeks: clarity of the core message under pressure, listening behavior, the quality of questions asked, decision hygiene under time constraints, and follow-through on commitments made in lab. Observers score at week one and week twelve, and we ask sponsors what they have noticed around week eight, which is usually when changes start showing up in real meetings rather than only in the room.

I want to be honest about the limits here, because this is where many REDIT programs oversell. Rubrics reward polish. A confident participant can score well while being genuinely difficult to work for, and a quieter participant can undersell changes that their team experiences as significant. So we triangulate: the transfer reflections, the sponsor conversations, and the participant's own account of what shifted in their real team carry as much weight as the observation scores.

We also refuse something, and sponsors occasionally push back on it. The labs do not produce scores for promotion or pay decisions. The moment a lab doubles as an assessment center, participants start performing safety. They choose gentle scenarios, they hedge in simulations, and the candor that makes rehearsal valuable disappears. When an organization needs assessment, that is a different product with different safeguards, and we would rather say so than run a compromised version of both.

Timing and selection matter more than content

The window where labs deliver the most is the transition itself, roughly three months before someone takes on people leadership to about a year after. Senior individual contributors preparing for a step sit comfortably alongside first-time leads; one group sees ahead, the other sees behind, and both profit from the exchange. For experienced managers with one specific gap, a full cycle is usually the wrong tool, and a short four-session format focused on that single theme works better.

Selection has edge cases worth respecting. There is the reluctant expert, like Daniel, a quality control specialist who accepted a lead role, in his words, "because otherwise they would have hired someone above me." He did not need confidence theater. What the lab surfaced was that he had never negotiated the actual scope of the role, and that conversation, with his sponsor, was worth more than any session on delegation. At the other end is the high performer sent as a reward rather than out of need. They tend to coast. We now ask sponsors to name the specific situation the participant will face in the next six months, and when there is none, we defer the seat to the next cohort rather than fill it.

On size, we have empirical regrets. A cohort of fourteen taught us that debrief quality collapses as the circle widens. Ten to twelve is the ceiling, and ten is better.

What we got wrong the first time

The early labs were over-scripted. We wrote scenarios with beats and expected lines, and participants dutifully delivered them while learning almost nothing, because nothing pushed back. We also tried to cover too much per session; cutting roughly a third of the material doubled the value of the debriefs, which should have been obvious from the start.

The biggest correction was removing sponsors from the live room. In our first cohort, a sponsor observed from the back row. Feedback turned polite, simulations turned performative, and we spent the remaining weeks rebuilding trust that one seating decision had cost. Sponsors now receive a written summary after each block and join a closing three-way conversation. Nobody has complained that they see less, and participants tell us it is why they can take risks.

The part that happens outside the room

Transfer is where programs quietly fail, and we design for it from day one. The cycle opens with a three-way conversation among participant, manager, and facilitator, agreeing on one real situation the participant will practice against. We steer people toward conversations whose worst realistic outcome is recoverable. A lab is the wrong place to rehearse a dismissal for the first time, but a good place to rehearse the preparatory conversation that precedes a formal process, with HR already involved.

Check-ins at week six and week twelve keep the transfer honest, because transfer assignments drift otherwise. And when the participant's own manager is part of the problem, we adjust the goals toward what the participant actually controls, and we tell the sponsor plainly where the limits sit. Overpromising there costs credibility that no curriculum can buy back.

If you are building something like this

A few lessons compressed, for anyone designing a similar program inside their own organization:

Start with one cohort of eight, not with a rollout plan. Harvest every scenario from participants' real work, anonymized and pressure-tested. Budget at least twice as much time for the debrief as for the simulation. Keep development and appraisal in separate rooms, permanently. Design the transfer assignment before you design the session.

Week twelve

At the end of her cycle, Sofia ran her conversation again, rebuilt with the real constraints her situation carried. This time she opened with