By Sahver Kaya · 6 min readA manager’s role in behavior change is to see what moved and coach from it. Practice builds the skill, but a change that only the person can see cannot be reinforced, redirected or recognized, and most behavior change never shows up anywhere. People sit through a workshop, nod, and go back to doing what they did before. Nobody can tell whether anything moved.
RCM ThinkLabs (rcmlabs.io) is an enterprise behavioral technology and decision-science company, and its product is a game-theory engine for measuring decision performance, built on advanced game theory (research at MIT with Prof. Muhamet Yildiz) and the behavioral science of Karl Kapp. It runs as a loop with two sides, one for the people practicing and one for the managers who coach them, and each side makes the other better.
Where behavior change disappears
The usual program ends at the person. They learn something, feel good about it, and walk back into a team that has not changed. The manager, who has the most influence over whether a new habit survives, is the person with the least information about it. They hear that the workshop was well rated and find out nothing else.
That is half of why one-off programs fail to stick. The other half is that behavioral skills fade without repetition. A change nobody sees and nobody reinforces is a change that does not last.
The loop, step by step
Practice. In short recurring sessions of up to fifteen minutes, people make real decisions inside a story with an incomplete picture. This is where critical AI-era cognitive skills get built, one rep at a time.
Feedback. When a session ends, a character from the story debriefs the person. RCM ThinkLabs has read how they listened, what they asked and how they decided, so the debrief is specific to them. Once a month they get a deep-dive report on how they are growing.
Coaching. This is the step most programs do not have. Managers can talk to RCM Advisor any time. Who on the team is avoiding hard questions this week? Why did someone’s reasoning shift? What should come up in Thursday’s one on one? Every 30 days, leadership gets a deep dive on the whole team.
Compound. Nothing resets between sessions, so every one starts higher than the last.
What a manager sees that they could not see before
RCM Advisor compares each person’s decisions with their own opening baseline. It tells a manager where movement is happening, where it has levelled off, and which conversation is worth having this week. It reads decisions, not opinions, so it does not depend on who spoke last or who speaks loudest.
It is built for development, not ranking. It supports the coaching conversation and does not stand in for it. A manager still has to know the person, and now they walk in knowing what the person actually did.
| Property | Workshop or course | RCM ThinkLabs Serious Games |
|---|---|---|
| What the person gets | An event, then nothing | Short recurring sessions and a private debrief from a character in the story |
| What the manager gets | Attendance and a satisfaction score | A read on how each person decides, through RCM Advisor, and a monthly team deep dive |
| What compounds | Nothing; each event starts from zero | Each session starts higher than the last, and the manager’s coaching raises the next one |
| Backing | Facilitator methodology or content library | Advanced game theory and behavioral science |
Early-career employees need the manager most
AI took much of the entry-level work that used to teach people how to think on the job, so early-career employees are building critical AI-era cognitive skills with fewer reps than any generation before them. We cover that gap on our page about the missing rung. Practice gives them reps. It does not tell them what to do with what they learn, and that is the manager’s part.
Many of those managers are new to the role themselves, coaching people who picked up AI faster than they did. We wrote about why a strong individual contributor struggles as a manager, and the loop is built for that moment. The manager does not need to out-prompt the team. They need to know how each person decides and have something specific to say next.
The two sides feed each other. The debrief gives the person something concrete to bring to a one on one, and the one on one tells them what to practice next. That is the design: managers grow into better coaches as the loop turns. We have not measured that yet, so we say it as intent, not as a result.
What the first deployment showed
Before the cycle began, one manager privately named a single dimension that held an engineer back. RCM ThinkLabs was never shown that note. Partway through the cycle it independently identified the same dimension, and once the read was confirmed and developed, that engineer’s signal on it rose 58%. Manager intuition and independent measurement pointed at the same gap, which is far more useful than either alone.
Across the team, over the same cycle, 143% more of the team’s questions asked why, up from 1 in 17 to 1 in 7, in 39 days. Among regular participants, 84% improved, at 70% voluntary daily engagement against a 5 to 25% norm. Those results describe participants, and they are on our evidence page with the rest of the production data.
“It’s working. It’s also creating a community, which is what we want.”Kristin Kupres · Chief Operating Officer, Wingbrace
Why the compounding is the point
A quarterly program starts from zero every time. A loop that keeps turning does not: practice recurs, nothing resets, and each rep lowers the cost of the next. By month six the person is reasoning against everything the first five months built, and the manager is coaching someone with a visible record instead of a recent impression.
Every deployment starts with an Initial Operating Capability, where each person enters at a measured baseline. That is the starting line for the compounding, not the finish. We lay out the path from baseline to mastery on the platform page, and if you want to see what the manager side would show for your own teams, start with a baseline.
Common questions
What is the manager’s role in behavior change? To see what changed and coach from it. Practice builds the skill, but a change nobody can see cannot be reinforced, redirected or recognized. RCM ThinkLabs gives the manager a read on how each person actually decides, through RCM Advisor, so the next conversation starts from what happened instead of a hunch.
Why does behavior change fade without the manager? People return to a team that has not changed. If the manager cannot see what a workshop moved, nothing at work reinforces it and the old habit wins. The skill also fades without repetition, so a one-off event with no coaching after it leaves nothing to compound.
How does RCM ThinkLabs help managers coach? Managers can ask RCM Advisor which people are moving, who is avoiding hard questions, why someone’s reasoning shifted, and what belongs in the next one on one. The answers come from decisions people made in the serious game, compared with each person’s own baseline, not from a survey or a recent impression.
How does the platform develop managers as well as early-career employees? Early-career employees get short recurring sessions and a private debrief from a character in the story. Their managers get the coaching read. Each side feeds the other: the debrief gives the person something specific to bring to a one on one, and the coaching tells them what to practice next. We designed the loop to make managers better coaches over time; we have not measured that yet.
What does the evidence show about behavior change? In production with an advanced engineering team, 84% of regular participants improved, at 70% voluntary daily engagement against a 5 to 25% norm. The share of the team’s questions that asked why rose 143%, from 1 in 17 to 1 in 7, in 39 days. These results describe participants, not managers.
See what the manager side would show for your teams.