The first management job is the only promotion in most careers where the new role has almost nothing in common with the one that earned it. Someone is promoted for being good at the work and then asked to stop doing the work, and whatever made them good at it is not what the new job needs.
It is also the level with the best evidence that development works and the least money spent on it.
Junior leaders change their behaviour more
The 2017 meta-analysis of leadership training, pooling 335 independent samples, tested whether the seniority of participants moderated outcomes.
Programmes delivered to low-level leaders produced significantly stronger transfer, meaning behaviour change back at work, than programmes delivered to high-level leaders and than programmes delivered to middle-level leaders. There was no significant difference between high and middle levels. On learning, no level differed significantly from any other.
So all three levels absorb the material equally well, and the junior group is markedly more likely to do something different afterwards.
Training programs provided to a sample of low-level leaders were associated with significantly stronger transfer effect sizes than those provided to high-level leaders or middle-level leaders.
The likely explanation is not that junior managers are better students. It is that they have fewer established habits to displace, less of an identity invested in how they currently operate, and a smaller gap between what a training scenario looks like and what their Tuesday looks like. An executive programme is often further from the participant's actual working life than a first-line management programme is from a team leader's.
Budgets are allocated in the opposite direction. Cost per participant rises steeply with seniority, and the residential executive programme is the most expensive line in most leadership budgets. The evidence does not say that spend is wasted; it says the same money one or two levels down is more likely to produce a change in behaviour.
The role got harder while nobody was looking
Gallup's State of the Global Workplace 2026 Report puts manager engagement at 22 percent in 2025, down from 27 percent in 2024 and 31 percent in 2022. Overall employee engagement moved one point in the same year, from 21 to 20 percent.
A five-point annual fall in one population while the surrounding population moves one point is a strong signal that something specific happened to that population. The report does not establish what, and the wider picture is covered separately. Four changes to the job are well documented and are all consistent with the pattern.
Wider spans. Delayering removes management layers without removing the people to be managed, so the remaining managers get more direct reports each.
Explicit coordination. Hybrid working converted a large share of team awareness from a by-product of proximity into scheduled work. Knowing how someone is doing used to be free.
Retained individual contribution. Most first-line managers still carry a personal workload. The management half of the job is the half with no deadline, which means it is the half that slips.
No support from above. Their own managers are in the same population at the same 22 percent.
The claim about talent, and how to read it
Gallup also publishes a figure that circulates widely: that about one in ten people possess the talent to manage, that 18 percent of people currently in management roles show a high level of talent for it, and that companies fail to choose the candidate with the right talent 82 percent of the time.
These come from the same proprietary database, measured across 27 million employees and more than 2.5 million work units over two decades, and the same caveat applies as to the claim that people leave managers rather than companies: the underlying data is not released for independent examination, and the construct of managerial talent is Gallup's own.
Treated as a measurement, the figure is unverifiable. Treated as a proposition, it is worth taking seriously anyway, because the mechanism behind it is not in dispute. Organisations routinely promote on the basis of individual performance in a role that shares few requirements with management, and there is no mystery about why that produces a poor hit rate.
The difference between those two readings matters for what you do. If one in ten people can manage, selection is the whole problem and development is triage. If the promotion criterion is wrong, then selection is fixable and the population of managers is not fixed at all. The second reading is at least as consistent with the data, and it is the only one that anyone can act on.
Who actually gets developed
The training data adds a dimension the engagement data does not.
In the EU in 2020, 42.4 percent of employees attended any training course during the year, and the figure falls with company size: 54.5 percent in enterprises with 250 or more employees, 27.5 percent in those with 10 to 49. Those figures cover all training, from compliance refreshers upwards.
A first-line manager in a mid-sized company is therefore in a population where fewer than two in five people receive any structured training in a year, and where a leadership programme specifically is a small fraction of that. The typical preparation for a first management job in a company of that size is a conversation, a handover, and a job title. That is covered in more detail in what firms of different sizes spend.
What the evidence supports doing
Four things, in order of how well supported they are.
Move budget down a level. This is the best supported conclusion in this article, from 335 samples, and it is a reallocation rather than an increase.
Train before the promotion, not after. No meta-analysis has tested timing directly, so this is an inference: the transfer advantage at junior levels is plausibly about the gap between the training scenario and the daily reality, and that gap is smallest at the point of transition.
Space it. Spaced programmes produced significantly stronger transfer and results than single-block programmes in the same meta-analysis. A first-line management programme delivered as one intensive day is the format the evidence supports least.
Include practice. Programmes combining information, demonstration and practice significantly outperformed those using information alone. The behaviours a new manager needs, giving difficult feedback, running a one-to-one that is not a status update, saying no to their own manager, are all things people are bad at the first three times they do them. The question is whether the first three times happen in a workshop or on a direct report. The full set of design findings is in what the evidence says leadership training changes.
Common questions
Does leadership training work better for junior managers? On behaviour change, yes. A meta-analysis of 335 samples found significantly stronger transfer for programmes delivered to low-level leaders than to middle or senior leaders. On how much participants learned there was no difference by level.
Why do junior leaders show more behaviour change? The research establishes the effect rather than the mechanism. The most plausible explanations are fewer established habits to displace, less identity invested in current ways of working, and a smaller gap between training scenarios and daily work.
Is it true that only one in ten people can manage? That figure comes from Gallup's proprietary database and cannot be independently verified. The underlying proposition, that promoting on individual performance selects poorly for management, is not in dispute and does not depend on the number.
How engaged are first-line managers? Gallup put manager engagement at 22 percent in 2025, down from 27 percent in 2024 and 31 percent in 2022, while overall employee engagement fell one point over the same year.
Should new managers be trained before or after promotion? No study has tested the timing directly. The transfer advantage at junior levels suggests that proximity between training and real situations matters, which favours training at or before the transition.
What format works best for first-time manager training? Spaced across weeks rather than delivered in one block, and combining information, demonstration and practice rather than any of the three alone. Both findings come from the same meta-analysis of 335 samples.
How much training does a typical employee actually get? In the EU in 2020, 42.4 percent of employees attended any course during the year, rising to 54.5 percent in large enterprises and falling to 27.5 percent in firms with 10 to 49 employees.
Sources
- Lacerenza, Reyes, Marlow, Joseph & Salas (2017), Leadership training design, delivery, and implementation: A meta-analysis, Journal of Applied Psychology 102(12)
- Gallup, State of the Global Workplace 2026 Report
- Gallup, Why Great Managers Are So Rare
- Eurostat, Participants in CVT courses as a percentage of persons employed (trng_cvt_12s)
