You Can Never Capture the Whole System

After years designing simulations such as an onboarding game for new hires at Nike, a team building game in maternity clinics for JHPIEGO, and a strategic planning game overcoming practitioner burnout at hospitals in South Africa for FPD. I've learned one hard truth: You never capture the whole system.

What you can capture—if you listen carefully—is just enough of reality to help people rehearse the choices they rarely get to examine safely.

That understanding fundamentally reshaped how I approach interviews and modeling. I no longer treat interviews as a quest for complete or definitive truth, but as a disciplined way of listening for enough of the system—for a specific learning purpose, at a specific moment in time, and from a specific set of perspectives. Even then, what emerges is never final. The model you build is reflective rather than authoritative, provisional rather than complete. Its role is not to provide answers, but to invite inquiry. What follows is a summary of the interview stages I’ve learned to use when building such models.:

Stage 1: Framing the System — What Are We Actually Trying to Understand? The first interviews are never about "content." They're about scope and intention.

At Nike, the question wasn't "How do we train managers?" It was: "What actually differentiates managers who succeed here from those who don't—given the culture as it really operates?"

At JHPIEGO, it wasn't "How do we teach clinical best practices?" It was: "Why do communication breakdowns persist even when everyone knows what they're supposed to do?"

At FPD, it wasn't "How do we reduce burnout?" It was: "How does burnout emerge across individuals, workplaces, and communities when everyone is already stretched?"

At this stage, I'm listening for boundaries: What's in scope? What's explicitly out of scope? What would make this work politically, culturally, or ethically unacceptable? Already, the model is forming—not from answers, but from constraints.

Stage 2: Manager and Leader Interviews — Where Strategy Meets Reality. Managers live in the tension between aspiration and execution. Their interviews often reveal the operating logic of the system.

At Nike, front-line managers consistently described success as improvisational. Formal tools existed, but what mattered was judgment—when to step in, when to get out of the way, when to protect the team, when to push. One manager described their role as "bringing it all together when plans fall apart." That sentence mattered. It told me that any simulation portraying management as linear or prescriptive would be dismissed immediately.

In the maternity clinic work with doctors, nurses, and midwives, senior clinicians revealed something different: hierarchy wasn't just structural—it shaped who felt permitted to speak in critical moments. The model couldn't simply reward "good communication." It had to represent why silence felt safer than speaking up.

At FPD, hospital managers spoke about impossible tradeoffs: staffing shortages, budget constraints, community expectations. Burnout wasn't accidental—it was baked into the system. Any model that framed burnout as a personal failure would be dishonest.

At this stage, I'm not trying to reconcile perspectives. I'm letting contradictions stand.

Stage 3: Front-Line Interviews — Where the System Shows Its Cracks. If you want to understand a system's failure modes, talk to the people in the trenches who absorb its consequences.

At Nike, front-line managers talked about learning through "critical experiences," not training programs. They described confidence built in moments where there was no playbook. As one manager put it: "In the time it takes to complete a training module, you could have worked the matrix and met someone who actually helps you advance." That single insight reshaped the entire design. Any simulation built on the assumption that managers learn through formal training would have failed.

In maternity clinics, nurses and midwives talked about moments where everyone knew what should happen, but timing, fear, or workload prevented coordination. Outcomes hinged on seconds and trust—not knowledge. One clinician said simply, "Everyone is doing their best, but the system doesn't let us talk when we need to."

In the hospital wellness work, nurses spoke about emotional detachment not as apathy, but as survival. One insight stayed with me: "The question isn't why people burn out—it's how they keep going at all."

This is where I have to be careful not to over-generalize. These interviews don't represent everyone. They represent enough voices to reveal patterns worth modeling.

Stage 4: Expert Interviews — Naming the Systemic Patterns. Experts help articulate what practitioners feel but may not name.

In the wellness work, a mental health expert reframed burnout as a systemic outcome rather than an individual deficit. She observed, "I often wonder how people manage to function under these conditions rather than why they don't." That reframing directly shaped the model: burnout became an emergent property of workload, leadership decisions, and social context—not a personal failure state.

In healthcare communication, experts helped identify how hierarchy and risk aversion interact—why even well-intentioned systems reproduce silence.

These interviews don't "validate" the model. They help me avoid building something naïve.

Stage 5: Building the Model — Choosing What Not to Include. This is the hardest and most uncomfortable stage. Every model is a reduction. Every simulation leaves things out. The question isn't whether it's incomplete—it's whether the omissions distort the learning purpose.

For The Next Wave, the model focused on tradeoffs between autonomy, alignment, and performance—not every aspect of Nike culture. Players navigate ambiguity the way real Nike managers do: improvising when plans fall apart, choosing when to protect their teams and when to push them.

For Maternity Clinic, we didn't model the entire health system. We modeled communication under pressure in a clinic context. The game surfaces how small delays, silence, or deference compound into serious patient risk. Communication choices—not medical facts—drive outcomes.

For The Wellness Wave, we didn't attempt to "solve" burnout. We modeled how it accumulates across roles and decisions. The game became a cooperative systems simulation where burnout emerges from structural pressures—staffing, leadership choices, community expectations—not from individual weakness.

At this point, I remind myself: This model is not reality. It's a lens. The diagram below becomes the backbone of the simulation. The game simply makes it playable.

Stage 6: Simulation and Game — Testing Recognition, Not Accuracy. When participants say, "That's not exactly how it works, but… yeah, that feels right," I know we're close. The goal is not fidelity to every detail. It's recognition. If players see themselves, their pressures, and their tradeoffs reflected, learning happens. If they don't, no mechanic will save the experience.

At JHPIEGO, practitioners consistently remarked that "the scenarios feel uncomfortably familiar." That's precisely the point of serious games grounded in reality.

At FPD, when healthcare workers reacted to early prototypes, they didn't debate mechanics—they debated realism. One participant said, "This isn't about people being weak. It's about choosing between bad options every day and carrying that home with you." That feedback confirmed we were modeling the right system.

Questions I Keep Asking Myself (and Maybe You Will Too)

--> What voices did we miss—and how might that matter?

--> What learning purpose justified this particular reduction of reality?

--> If someone challenged this model, where would they be right?

Those questions never go away. And I don't think they should. Because the moment I believe I've captured the system, I've probably stopped listening. I'd genuinely welcome your thoughts. This work is always evolving, and the best insights often come from the challenges others are wrestling with.