What causes leadership development programs to fail is a small set of repeatable, structural reasons: the learning has nowhere to land once the workshop ends, there's no reinforcement in how the business actually operates day to day, the wrong people were selected for the wrong reasons, and success was never actually defined before the program started. Content quality is rarely the cause, which is the uncomfortable part, because content quality is the thing most businesses focus on fixing when a program doesn't work.
Reason One: The Learning Has Nowhere to Land
A capable leader can leave a genuinely excellent workshop having learned something real and directly relevant to them, and still have nowhere to apply it, because nothing about their actual role, decision rights, or daily cadence changed alongside the learning. They go back to the same meetings, the same unclear ownership, the same absence of a structure that would let the new skill actually get used. The learning doesn't fail because it was forgotten. It fails because it was never given anywhere real to go, and skills that have nowhere to go atrophy within weeks regardless of how well they were originally taught.
The Test: Ask a leader who finished a program six months ago what changed about how they actually, concretely work day to day, specifically, not how they feel about leadership in general. If the answer is vague, the learning had nowhere real to land.
Reason Two: No Reinforcement in How the Business Actually Runs
This second reason is closely related to the first one but is genuinely distinct enough to be worth naming separately in its own right. Even when a leader does have somewhere to apply new learning, the business's actual incentives, what gets rewarded, what gets promoted, what gets quietly tolerated, often pull in the opposite direction from what the program taught. A program can teach delegation while the business still promotes people specifically for doing everything themselves. It can teach difficult conversations while managers who avoid them face no real consequence. The program and the business end up teaching two contradictory lessons at once, and the business's day-to-day reinforcement wins almost every time, because it's constant and the workshop was a single week.
- No Landing Place: The role, decision rights, and daily cadence don't change alongside the learning, so new skills have nowhere real to be applied.
- Contradictory Reinforcement: What actually gets rewarded and promoted in the business pulls against what the program taught, and daily reinforcement usually wins.
- Wrong Selection: Attendees chosen by seniority, availability, or as a consolation prize, not because of a genuine, identified capability gap.
- No Defined Success: Nobody agreed in advance what changing would actually look like, so there's no way to know afterward whether it worked.
I've genuinely sat in enough leadership offsites over the years to notice a very consistent pattern in how businesses respond when a program clearly hasn't worked. The instinct is almost always to change the curriculum, bring in a different facilitator, add a module on accountability, refresh the slides. What almost never happens is a hard look at whether the business's own daily behaviour was ever going to let the new approach survive contact with how promotions, meetings, and rewards actually work. Changing the content is the easy fix. Changing the reinforcement is the one that actually matters, and it's the one businesses reach for last, if at all.
Reason Three: The Wrong People Were Selected
Program selection is genuinely where I see some of the most avoidable, and frankly most frustrating, failures happen in practice. Seats get filled by seniority, by who happens to be available that quarter, or, worse, as a quiet consolation prize for someone who was passed over for a promotion. None of those selection criteria have anything to do with whether the person has a genuine, identifiable capability gap the program is actually designed to close. A brilliant program delivered to the wrong audience produces glowing feedback scores and zero durable change, because the people in the room didn't need what was being taught, or needed something the program was never designed to address.
There's one specific version of poor selection genuinely worth naming carefully, because it's more common than the obvious cases: sending someone to a program as a way of signalling investment in them, without any real diagnosis of whether the content matches a gap they actually have. It comes from a genuinely good instinct, wanting to develop people, but it produces the same failure as any other mismatched selection. The person attends, engages politely, and returns to a role where the content doesn't map to anything they were actually struggling with, because the selection was about the relationship, not the diagnosis.
Reason Four: Nobody Defined What Success Actually Looks Like
This is the quietest failure and the one that makes all the others invisible. Without a specific, concrete definition of what changing looks like, agreed before the program starts, there's no way to know six months later whether it worked. The business defaults to feedback scores, how much people enjoyed the sessions, as a proxy for actual impact, and feedback scores measure the experience of attending, not whether anything about how people lead actually changed. A program can score highly on feedback and change nothing measurable, and without a defined success criterion nobody in the business will ever notice the gap between those two things.
Defining success has to genuinely happen before the program actually starts, not after the fact, because retrofitting a definition once the program is already running almost always produces something vague enough to be technically satisfied regardless of what actually happened. A real definition looks like a specific behaviour change, tied to a specific role, observable by a specific person, within a specific timeframe, not a general sense that leadership feels stronger. Vague goals produce vague evaluations, and vague evaluations are how mediocre programs get renewed year after year without anyone quite being able to explain why.
- Define success before the program starts — Agree on the specific, observable behaviour change you're expecting, tied to a role and a timeframe, not a general feeling that leadership improved.
- Select attendees by capability gap, not availability — Every seat should map to a genuine, identified gap the program is designed to close, not seniority or who happened to be free that quarter.
- Change something structural alongside the learning — Give the new skill somewhere real to land, a changed decision right, a new responsibility, or the learning has nowhere to be applied.
- Check what the business actually reinforces — If promotion and reward criteria contradict what the program teaches, fix the reinforcement first, or the program is fighting a battle it can't win.
I'd genuinely add one more layer to properly defining success here, because even businesses that try to define it often stop one step short. It's not enough to name the behaviour you're expecting to change. You have to name who's going to observe it and how, before the program starts, or the evaluation quietly becomes retrospective and self-serving. A defined success criterion with no observer and no method is really just a hope written down more formally, and hopes written down formally still don't tell you, six months later, whether anything actually changed.
Why It Causes Leadership Development Programs to Fail Even in Sophisticated Businesses
None of these four causes genuinely require a poorly run or badly managed business to occur, which is why they show up even in genuinely sophisticated companies with real leadership development budgets. It's much easier to evaluate a program by its content, the quality of the facilitator, the design of the curriculum, than by whether the surrounding structure was ever set up to let that content actually land. Content is visible and easy to judge in the room. Structural readiness is invisible until months later, when the absence of change finally becomes undeniable, by which point the program has usually already been renewed for the following year on the strength of good feedback scores.
It's also genuinely worth being honest here that none of this is an argument against leadership development itself, as a category. Some of the strongest capability shifts I've seen in leaders came directly out of a well-designed program, delivered to the right person, at the right moment, with a clear place to apply what they learned. The point isn't that programs don't work. It's that whether they work depends far more on the four structural conditions around the program than on the program's content, and businesses that keep investing in better content while ignoring the structural conditions will keep getting the same disappointing result, just with better production values each time.
The Distinction That Actually Matters
Leadership development programs rarely fail because the content was weak. They fail because the business around the program was never actually built to let the learning land, the incentives quietly pulled against it, the wrong people were in the room, or nobody defined what success meant before the program began. Fix those four structural conditions first, and even a modest program starts producing real, durable change. Skip them, and even an excellent one will keep generating great feedback scores and nothing that actually lasts.
Before commissioning the next program, or renewing the current one purely on the strength of last year's feedback survey, it's worth asking the four questions directly: where will this learning actually land, what does the business currently reward that might contradict it, who genuinely needs this and why, and what, specifically, will count as proof it worked. Answer those honestly before the budget gets spent again, and the program has a real chance this time.