Why most leadership programmes fail
The common failure is not bad content. It is that the programme is an event rather than a system. People attend two days of training, return to a manager who was not involved and a job that has not changed, and within a month the behaviour reverts.
Research on training transfer consistently finds that what happens before and after the classroom matters more than the classroom. A programme with a mediocre curriculum and strong reinforcement outperforms an excellent curriculum with none.
The sequence that works
- Define the behaviour, not the topic. “Managers run weekly one-to-ones that are not status updates” is buildable and measurable. “Improve leadership capability” is not.
- Find where the gap actually is. Pull turnover and engagement by individual manager rather than by department. It clusters, and the clustering tells you who to develop.
- Involve the manager above. The single strongest predictor of transfer. If a participant’s manager never asks about it, the programme signals that it is optional.
- Space it out. Six sessions over three months beats two consecutive days, because the practice between sessions is where change happens.
- Make practice compulsory and specific. One behaviour, one context, reported back.
- Measure the behaviour. Not satisfaction scores. Whether the one-to-ones happen and whether reports say they changed.
What to include
For a first-time manager cohort the highest-return content is consistent: delegation, one-to-ones, difficult conversations, and accountability. Those four cover most of what new managers get wrong.
For experienced managers moving toward senior roles, the content shifts to influence without authority, executive presence, and decision-making under uncertainty.
Build, buy, or blend
| Strength | Weakness |
|---|
| Build internally | Fits your context; cheap to repeat | Needs internal capability and time |
| Buy a programme | Fast, proven curriculum | Generic; transfer still your problem |
| Individual coaching | Highest impact per person | Does not scale economically |
| Blend | Group for common ground, coaching for the few who most affect outcomes | Requires coordination |
Most organisations under about 200 people are better served by the blend than by building a formal curriculum.
Frequently asked questions
How do you build a leadership development programme?
Start from the specific behaviours you want, not topics. Identify where the gap actually sits by looking at turnover and engagement by individual manager. Involve participants’ own managers, space sessions over months rather than compressing into days, require specific practice between them, and measure behaviour rather than satisfaction.
Why do leadership training programmes fail?
Because they are events rather than systems. Participants return to an unchanged job and an uninvolved manager, and behaviour reverts within weeks. What happens around the training matters more than the training.
How long should a programme run?
Three to six months with sessions every two or three weeks. Compressed two-day formats produce good satisfaction scores and little durable change, because there is no space for practice.
How do you measure whether it worked?
Define the target behaviour before starting and measure that specific thing — whether one-to-ones happen, whether reports describe feedback differently, whether turnover under those managers moves. Satisfaction scores measure enjoyment, not transfer.