Companies spend roughly $371 billion a year globally on corporate training (Training Industry Report, 2023), and a substantial fraction goes to what the L&D field calls soft skills: leadership, communication, collaboration, emotional intelligence, negotiation, empathy, decision-making under uncertainty. These are the skills executives rate as most important for organizational performance and the skills the same executives consistently say their workforce is worst at. They are also the skills the training industry has the hardest time producing durable gains in. The gap between what gets spent and what actually transfers is well-documented and embarrassing.
RPG mechanics solve a specific piece of this problem, because the mechanics of a well-designed RPG happen to operationalize the exact three conditions the learning science literature identifies as necessary for soft-skill transfer.
The 10% Problem: Why Workshops Don't Stick
Start with Hermann Ebbinghaus, the German psychologist who in 1885 ran the first rigorous experiments on human memory. Ebbinghaus memorized nonsense syllables and tracked how much he remembered at intervals. His curve showed memory decays sharply after initial exposure: roughly 50% forgotten within an hour, 70% within a day, less than 10% retained after a week without reinforcement. The curve has been reproduced thousands of times across different populations, materials, and contexts. It is one of the most robust findings in cognitive psychology.
Corporate soft-skills training runs head-on into this curve. A typical leadership workshop is a one-to-three-day event. Participants absorb a large volume of content. They go back to work. Within a week, by the numbers, retention is single digits. Within a quarter, it is effectively zero. The training industry knows this. McKinsey's research on training effectiveness consistently finds behavior-change gains in the low single digits relative to training spend.
Why does this happen for soft skills specifically? Two reasons. First, soft skills are behavioral skills, not declarative knowledge. Knowing the four styles of conflict resolution is not the same as doing the harder of the four styles when your heart rate is elevated. Declarative knowledge decays on the Ebbinghaus curve. Behavioral skill requires something different.
Second, the transfer context is dissimilar from the training context. Training happens in a conference room with a facilitator and a clear correct answer. Real use happens at a desk with a hostile email and no facilitator and no correct answer. Research on transfer of training in Academy of Management Learning & Education has long documented that similarity between training and transfer contexts is the strongest predictor of transfer occurring. Conference-room practice does not resemble the situations where the skill has to be deployed.
Anders Ericsson and the Three Conditions for Skill
The alternative framework comes from Anders Ericsson's research on expertise. Ericsson spent decades studying how people become world-class at complex skills: chess masters, concert pianists, Olympic athletes, top-tier surgeons. His 1993 paper with Krampe and Tesch-Römer in Psychological Review introduced the concept of deliberate practice, and it remains the foundational framework in the expertise literature (Ericsson, 2008, Academic Emergency Medicine, covers the updated synthesis).
Ericsson identified three conditions for skill acquisition beyond what lectures can provide:
- Focused, repetitive practice of the specific target skill. Not reading about the skill. Not watching someone else do it. Doing it, repeatedly, at the edge of your current ability.
- Real-time, constructive, actionable feedback. The person practicing needs to know what they just did and whether it worked, fast enough to adjust.
- Performance at the edge of current competence. The task needs to be hard enough that the learner can't do it automatically but easy enough that they can do it with effort.
Subsequent research has added a fourth condition for skills that need to transfer across contexts: varied conditions. Pilot simulators train for multiple airports, weather conditions, equipment failures. Surgical simulators train for anatomical variants and complications. The varied conditions force the learner to abstract the skill from specific instances and develop a generalizable capacity.
Now compare this to a typical corporate leadership workshop. The workshop provides exposure to concepts. It does not provide repeated practice at the edge of current ability. It does not provide real-time feedback on behavioral attempts. And it does not provide varied conditions. It provides one-time exposure to information. Ebbinghaus's curve takes care of the rest.
Why RPG Mechanics Fit the Research
RPG mechanics, when designed for training, operationalize Ericsson's four conditions directly.
Repetitive practice. A well-designed RPG scenario puts the player in a decision point multiple times per session. Over a quarter of weekly sessions, the player has faced dozens of situations requiring the target behavior. That's dozens of repetitions compared to the one-to-two repetitions a workshop provides.
Edge of current ability. The adaptive difficulty of a good RPG places the challenge near the player's capability frontier. The scout character can solve the easy problems on their own. The hard problems require the player to try something they haven't done before. The friction is the training.
Real-time feedback. The game provides immediate consequences. A dissent that changes the team's direction gets recognized. A prosocial sacrifice registers. A failure to speak up when the team is going the wrong way produces a visible outcome. The feedback loop is tight enough that the learner can calibrate.
Varied conditions. Each session is a different scenario. Different resources, different antagonists, different team configurations. The underlying skill (voicing a concern, navigating task conflict, committing to a risky plan) has to be redeployed in different contexts. That's exactly the varied-conditions requirement the transfer research identifies.
Compare this to a workshop. One scenario. No repetition. No consequences. No variation. The research predicts the observed outcome.
The Specific Soft Skills RPGs Teach Well
Not all soft skills are equally amenable to RPG-based training. The ones with the strongest fit share three features: they are behavioral (observable action, not just understanding), they are interpersonal (require another person in the loop), and they require calibration under uncertainty (no scripted correct answer).
Leadership under ambiguity. The scenario gives the team incomplete information and requires a decision. The player who takes the lead, proposes a plan, and commits the team to it is practicing the exact skill managers describe as the hardest and most valuable part of their job. See Best Leadership Courses for New Managers for how this skill has historically been trained and where the gaps are.
Communication under pressure. The team has to coordinate in real time to handle a situation that is deteriorating. Who speaks, who listens, who summarizes, who decides. These are micro-behaviors that compound into team communication patterns. RPG sessions surface the patterns and produce the repetitions.
Negotiation. When characters have conflicting goals and shared resources, they have to negotiate. Unlike a conference-room negotiation exercise, the stakes are real within the fiction, which produces the emotional arousal that encodes durable memory (Tyng et al., Frontiers in Psychology, 2017). Negotiation in an RPG is practiced, not discussed.
Empathy. Playing a character whose situation differs from the player's own is a structured empathy exercise. Batson's 1991 research on perspective-taking showed that imaginative perspective-adoption produces measurable empathy gains. RPG play is perspective-adoption with mechanical reinforcement across multiple session.
Decision-making under uncertainty. Kahneman and Tversky's 1979 prospect theory identified loss aversion as the dominant cognitive bias in risky decision-making. Real decision-making under uncertainty requires calibrating the loss aversion against the opportunity cost of playing safe. RPG scenarios with gamble-or-safe mechanics (covered in depth in How to Build Psychological Safety Through Play) produce hundreds of these decisions over a session.
Harvard, MIT, and the Evidence for Experiential Learning
The effect sizes for experiential and simulation-based learning in rigorous research are large enough to be worth naming specifically.
Harvard Business School's FIELD program, launched in 2011, embedded experiential exercises throughout the MBA curriculum. Students practice soft skills through activities including identity mapping and simulated team decisions, in what the Harvard Office of the Vice Provost for Advances in Learning describes as low-stakes, experimental ways to practice communication and team effectiveness before deploying the skills in real work. The program's internal assessment documented gains that traditional case-method teaching had not produced.
MIT Sloan's Namrata Kala ran a randomized controlled trial across five factories in Bangalore on soft-skills training. The results, published in the Quarterly Journal of Economics, showed that in-factory soft-skills training returned roughly 250% ROI within eight months of its conclusion. The training delivered measurable behavior change. Note what the study controlled for: the training was experiential and delivered in context, not conference-room workshop style.
PwC's 2020 study on VR-based soft-skills training found VR training was four times faster than classroom-based training and increased confidence in applying acquired skills by 275%. The mechanism is the same one the RPG literature points to: immersive simulation produces emotional engagement, which produces memory consolidation, which produces retention and transfer.
The 70-20-10 model used by corporate L&D (70% of learning from experience, 20% from social interaction, 10% from formal training) captures the same insight at a budget-category level. The research does not say formal training is worthless. It says the 10% that comes from training is only useful if it is coupled to experiential practice and social learning. RPG-based training is explicitly the experiential and social layer, which is why it complements rather than replaces the informational layer.
The Transfer Evidence from Adjacent Fields
Three adjacent fields have much stronger empirical records than corporate L&D, and they all share a common feature: simulation training is the core methodology.
Aviation. Commercial pilot training has used flight simulators for decades. The transfer-of-training research on aviation simulators documented effect sizes around 1.0 (enormous by the standards of education research) for specific skill transfer. Airlines don't train pilots in classrooms for the same reason they don't learn to drive by watching slideshows.
Medicine. Surgical simulation has become standard practice for high-stakes procedures. A meta-analysis of 14 studies on simulation-based medical education with deliberate practice found an effect size of 0.71 compared to traditional clinical training. Surgeons trained on high-fidelity simulators have demonstrably better patient outcomes than surgeons trained only in the traditional see-one-do-one-teach-one model.
Military. Combat simulation research documents similar effect sizes with similar mechanisms. The RAND Corporation's work on collective simulation-based training in the U.S. Army demonstrates that psychological fidelity matters more than physical fidelity for collective cognitive tasks (like synchronizing action with teammates). The insight: for team skills specifically, you don't need the simulation to physically resemble the work environment. You need it to produce the same cognitive and emotional demands. That is exactly what a narrative RPG does for team dynamics.
Corporate soft-skills training lacks the simulation layer that these adjacent domains have spent decades building. The result is the effect-size gap: medical and aviation simulation produce transfer, corporate workshops do not. RPG-based training is the simulation layer for the team-dynamics domain.
Where QuestWorks Fits
QuestWorks is purpose-built for the soft skills the research identifies as most amenable to RPG-based transfer. The mechanics are designed around the specific behaviors that separate high-performing teams from mediocre ones.
The dissent mechanic builds voicing-contrary-positions-under-social-pressure. The risk-and-gamble mechanic builds decision-making-under-uncertainty. The prosocial sacrifice mechanic builds commitment-to-the-group. The shared-fate structure builds interdependence and collective efficacy. Each mechanic is grounded in a specific body of organizational behavior research and operationalized into a game mechanic that produces repetitions under varied conditions. The full articulation is in Why Games Work for Team Development and the related pieces on each specific mechanic.
The product runs as a voice-controlled cinematic simulation on the QuestWorks platform. It integrates with Slack for install, invites, leaderboards, and a private AI coaching layer that employees can use for one-on-one conversations that stay private. The learning experience happens in the game. Slack is the infrastructure.
The Flight Simulator Frame
The clearest way to hold this is by the aviation analogy. Commercial airlines do not teach pilots to handle engine failures by sending them to a three-day workshop on engine-failure theory. They put pilots in flight simulators and fail their engines, repeatedly, in varying conditions, until the pilot's response is encoded at a level below conscious thought. Then when the real engine fails, the pilot's hands know what to do.
Soft skills on teams work the same way. The hard conversation, the moment where dissent has to be voiced, the decision to commit the team to a risky plan, the negotiation with a peer whose goals conflict with yours. These are the behavioral engine failures. The team that has rehearsed them fifty times in low-stakes RPG scenarios has the muscle memory when the real moment arrives. The team that has sat through a leadership workshop does not. That is the soft-skills version of the flight simulator for team dynamics, and it's the category gap RPG mechanics fill.
For broader context on why play works for team development generally, see Why Games Work for Team Development. For the historical comparison of what leadership training has looked like and why it keeps failing, see Best Leadership Courses for New Managers. For the specific mechanism by which psychological safety is built through play, see How to Build Psychological Safety Through Play.
What This Means for L&D Budgets
The practical implication is a reallocation question. If the research on Ebbinghaus, Ericsson, and training transfer is correct, then soft-skills budget allocated to one-time workshops is producing almost no durable behavior change. The same budget allocated to experiential training with repeated practice under varied conditions will produce substantially more. The question for L&D leaders is whether the existing budget mix reflects that evidence.
Most L&D organizations have not made the reallocation, because the infrastructure for experiential soft-skills training has not existed at scale. Workshops are cheap to procure and easy to schedule. Flight simulators for team dynamics did not exist. That is the gap RPG-based serious games fill, and it's why the category is growing at the rate that it is.
The soft-skills research and the training-transfer research have been pointing at the same answer for forty years. The adjacent fields (aviation, medicine, military) made the pivot to simulation decades ago and produced the effect sizes to validate the decision. Corporate L&D is overdue for the same transition. RPG mechanics, applied with care to the specific behaviors teams need to develop, are the most scalable version of that transition available in 2026.
Try it free
