Corporate learning is having a confidence crisis. Firms pour money into learning management systems, mandatory modules, and gamified platforms, and then watch completion rates limp and knowledge evaporate within weeks. The usual diagnosis is that the content is boring and the solution is better production values. The motivation research suggests the problem is deeper and more structural. A great deal of corporate training is designed, almost perfectly, to kill the motivation it depends on.
The three mistakes built into most programs
Start with rewards and mandates, the default machinery of L&D. Completion badges, points, leaderboards, and the implicit threat behind compliance training are all controlling incentives. Deci, Koestner and Ryan (1999), in a meta-analysis of 128 experiments, found that tangible rewards offered for engaging with, completing, or performing a task significantly undermined intrinsic motivation, with the effect strongest on tasks people already found interesting. A learning platform that gilds every module with points is not adding motivation. It is teaching people that the content is only worth touching for the badge, and quietly draining the intrinsic interest that makes learning stick.
The design flaw is not limited to rewards. Ryan and Deci (2020) argue that motivation of the highest quality, the internalized kind where a person actually endorses what they are learning, depends on supporting three needs: autonomy, competence and relatedness. Mandatory training violates autonomy by definition. One-size modules ignore competence, wasting the expert's time and losing the novice. And self-paced e-learning taken alone at a desk strips out relatedness. Miss all three, and no amount of slick video will save the outcome.
What actually moves the needle
The constructive evidence is specific. Wang and colleagues (2024) synthesized 36 intervention studies covering 11,976 participants and found that programs deliberately designed to support autonomy, competence and relatedness improved motivation. This was the first systematic review in more than a decade to test all three needs together, and it reframes the design problem. The lever is not engagement gimmickry. It is need support baked into the structure of the learning.
Autonomy support in particular pays off broadly. Mammadov and Schroeder (2023), drawing on 378 effect sizes from 179 samples across 153 studies, found autonomy support associated with six families of positive outcomes, including autonomous motivation, behavioral engagement, mastery goals, and self-regulated learning, with an average effect size of 0.32 (Mammadov and Schroeder, 2023). Self-regulated learning is exactly what a modern employer wants: people who can direct their own upskilling without being marched through it. And note the honest limit in that same study, the effects were larger on motivation and engagement than on hard performance measures. Autonomy support reliably makes people want to learn and take ownership of it. It is not a magic switch on next quarter's output, and no one should sell it as one.
Redesigning L&D around control, not content
The practical reframe is to stop asking "how do we make this course more engaging" and start asking "who controls this learning." Give people genuine choice over what they learn and how, so autonomy is real rather than decorative. Sequence for competence so that no one is thrown into material they cannot yet handle, since Deci, Koestner and Ryan (1999) found that feedback conveying competence, unlike controlling rewards, enhances motivation. Build relatedness back in through cohorts, mentors and peer teaching, so learning is a social act rather than a solitary compliance chore.
None of this requires a bigger content budget. It requires ceding control, which is the one thing centralized L&D functions are structurally reluctant to do. The irony is sharp. Organizations say they want employees who take charge of their own growth, then build learning systems that punish exactly that instinct with mandates and points. If your training teaches people to wait for the next assigned module, do not be surprised when they wait for the next assigned task too. The skill you most want, self-directed learning, is not a topic you can add to the catalog. It is a property of how the whole system distributes control.
Sources
- Deci, Koestner and Ryan (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. 128 experiments showing tangible rewards undermine intrinsic motivation while competence feedback enhances it, the core critique of gamified L&D. https://doi.org/10.1037/0033-2909.125.6.627
- Ryan and Deci (2020). Intrinsic and extrinsic motivation from a self-determination theory perspective. Framework showing internalized motivation depends on autonomy, competence and relatedness, all violated by mandatory one-size training. https://doi.org/10.1016/j.cedpsych.2020.101860
- Wang et al. (2024). SDT-based educational interventions: a meta-analysis. 36 studies and 11,976 participants showing need-supportive design raises motivation, the constructive template for L&D. https://www.sciencedirect.com/science/article/abs/pii/S0023969024000572
- Mammadov and Schroeder (2023). A meta-analytic review of the relationships between autonomy support and positive learning outcomes. 378 effect sizes, average 0.32, showing autonomy support drives self-regulated learning, larger on engagement than on hard performance. https://doi.org/10.1016/j.cedpsych.2023.102235