Return-to-office edicts hog the headlines, but the quieter crisis is that too few people feel like learning at work. Global employee engagement slid to 20 percent in 2025, the lowest since 2020, with Gallup estimating a $10 trillion productivity drain, roughly 9 percent of world GDP (Gallup, 2025). Managers are not immune. Their engagement has fallen 9 percentage points since 2022, including the steepest single-year drop from 27 to 22 percent between 2024 and 2025 (Gallup, 2025). When the very group meant to assign courses and track completions is running on empty, more scheduling and compliance modules will not fix the problem.
What people learn sticks when they choose it
Self-directed education is not a perk for knowledge workers with spare time. It is a way to align human motivation with the pace of change. The strongest theoretical and empirical scaffold for this idea comes from Self-Determination Theory, which holds that motivation deepens when three basic psychological needs are met: autonomy, competence, and relatedness. A new meta-analysis by Wang and colleagues synthesized 36 SDT-based educational interventions with 11,976 participants, selected through a PRISMA process from 505 records across nine databases. It is the first systematic review in more than a decade to assess interventions that address all three needs in education, and it explicitly tests whether such designs improve intrinsic motivation and related outcomes (Wang et al., 2024). This matters for companies because most corporate “learning” treats autonomy as a scheduling problem, competence as content delivery, and relatedness as a chat in the LMS. SDT invites a redesign: give people real choice over what and how they learn, ensure frequent, competence-building practice tied to real work, and build communities that make progress visible and social.
The earlier literature often focused on a single ingredient, like autonomy support. By contrast, Wang et al. survey the full recipe. Leaders should read that as a design brief. If your learning systems cannot show where autonomy, competence, and relatedness live in the day-to-day, you are not set up for durable motivation (Wang et al., 2024).
Psychological safety is the hinge
Even with autonomy, people will not choose stretch goals if it feels risky to admit they do not know. That is where psychological safety enters. In a field study of 51 teams in a manufacturing company, Amy Edmondson showed that team psychological safety is associated with learning behavior, while team efficacy is not once safety is accounted for. Learning behavior, in turn, mediates the link between safety and performance (Edmondson, 1999). The practical reading is blunt. If your teams do not feel safe to ask naive questions, surface errors, and experiment in public, self-directed education stalls. People will pick what looks easy or quietly opt out.
Safety is not a poster on a wall. It is what colleagues do when someone tries a new technique and it fails. It is whether senior leaders ask for early drafts rather than perfect slides. It is whether review rituals probe the process, not just the outcome. These are managerial choices. They are also the preconditions for any system that expects adults to own their learning.
Decentralize authority if you want adaptive learning
Companies serious about self-directed education should study how self-managing organizations distribute power. Lee and Edmondson define self-managing organizations as those that decentralize authority in a radical, formal way across the whole firm, eliminating the manager–subordinate relationship as the default. Their synthesis argues that conventional managerial hierarchy works well for reliable execution of known tasks in stable contexts, but it inhibits cross-functional problem solving, hardens rigidity, and reinforces status differences that stifle development (Lee and Edmondson, 2017). In other words, the very structure that makes instruction efficient can make learning brittle.
You do not need to abolish all managers to apply the lesson. Treat learning choices the way self-managing firms treat operational decisions: push decisions as close to the work as possible within clear constraints. That means publishing strategic skills your company needs, then letting teams and individuals propose projects that build those muscles. It means replacing a centralized course catalog with a visible marketplace of stretch assignments, mentorship offers, and peer-led circles. Authority in this system sits with workers and teams to choose learning paths, and with leaders to set direction and remove impediments. The logic echoes the evidence: decentralize for complex, non-routine problems, and your people will develop while solving them (Lee and Edmondson, 2017).
The manager’s new job
Gallup’s numbers tell us managers are stretched and often disengaged themselves (Gallup, 2025). Expecting them to micromanage learning on top of everything else is a design error. Redefine the role. Managers set outcomes, codify norms that protect candor, and make time and budget visible. They curate peer review rather than gatekeep access. They publish progress and invite help. Most of all, they model uncertainty. In Edmondson’s terms, they make risk-taking in learning feel safe because they do it too (Edmondson, 1999).
A workable blueprint looks like this. Start with a 90-day learning sprint in three teams working on real business problems. For autonomy, let individuals pick a focus skill and a project that uses it. For competence, require weekly practice with feedback, not just reading or videos. For relatedness, create small peer circles that meet biweekly to show drafts and trade help. Managers agree to time blocks for learning work and to public demos every three weeks. Capture what is learned in short write-ups that others can reuse. Throughout, leaders ask the questions that surface mistakes early and credit the teams that made them useful. The SDT frame keeps motivation on track, and psychological safety keeps the engine from seizing.
What to measure and why it matters
Skip vanity metrics like course completions. Track learning behaviors and business-relevant signals. Edmondson’s work highlights behaviors such as speaking up, seeking feedback, and experimenting. Make those observable in retrospectives and reviews (Edmondson, 1999). Pair that with tangible outputs from the 90-day sprint, such as decommissioned waste in a process or a faster customer response. If the patterns are positive, widen the circle. If not, adjust the design: add choice, increase practice, tighten the peer loop. That is the point. You are testing a management system, not hunting for the perfect course.
Leaders often ask whether self-directed education scales. The better question is whether your current system scales disengagement. With only one in five employees engaged and managers’ own energy ebbing, the status quo is not neutral, it is a drag on performance (Gallup, 2025). The research base gives you the contours of an alternative. Design for autonomy, competence, and relatedness in the learning experience (Wang et al., 2024). Build team climates where trying, erring, and asking are safe because they are normal (Edmondson, 1999). Push authority over how learning happens closer to the work so development and problem solving reinforce each other (Lee and Edmondson, 2017). Then treat the system itself as a continuous experiment. That is how adults learn, and how organizations do too.
Sources
- Wang et al. (2024). SDT-based educational interventions: a meta-analysis, Learning and Motivation 87. Synthesizes 36 interventions with 11,976 participants and tests whether designs that support autonomy, competence, and relatedness improve intrinsic motivation and related outcomes. https://www.sciencedirect.com/science/article/abs/pii/S0023969024000572
- Edmondson (1999). Psychological Safety and Learning Behavior in Work Teams, Administrative Science Quarterly 44. Field study of 51 teams showing psychological safety is associated with learning behavior, which mediates the link to performance. https://journals.sagepub.com/doi/10.2307/2666999
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing, Research in Organizational Behavior 37. Defines self-managing organizations and argues hierarchy suits routine work while inhibiting complex problem solving and development. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Gallup (2025). State of the Global Workplace 2025. Reports global engagement at 20 percent, a $10 trillion cost equal to 9 percent of GDP, and a 9-point drop in manager engagement since 2022. https://www.gallup.com/workplace/659610/state-global-workplace.aspx