What Game Design Taught Me About Commercial Systems

On nights and weekends, away from any commercial brief, I build worlds. Interactive ones, with their own rules, factions, economies, and ways to win. It started as play, a place to make things with no business case attached. What I did not expect was how much it would teach me about the work I do during the day.

The surprising part of building a world is what turns out to be hard. Not the art. Not the code. The hard part is behavior. Will anyone do the thing you designed for them to do? You can build something elegant and balanced on paper, then watch people ignore the path you intended and exploit the one you did not. A world is a machine for producing behavior, and behavior does not care how clever the design is. It responds to incentives, friction, feedback, and reward, roughly in that order.

So you learn to design for how people actually act. An incentive pointed slightly wrong will quietly bend the whole system toward the wrong outcome. Progression matters more than features: people stay when they can see the next step and feel themselves getting closer to it. Feedback has to be fast and legible, or people cannot tell whether what they did worked, so they stop doing it. And the most powerful lever is rarely the obvious one. Change a single reward by a little, and the entire pattern of behavior reorganizes around it.

The same problem, different clothes

These are the exact problems I spend my days on in commercial work. Why does a sales rep push one product and ignore another that is better for the customer? The incentive points somewhere else. Why does a launch lose momentum a month in? The feedback loop went quiet, nobody could tell it was working, and people drifted back to what they knew. Why does a partner sell your competitor instead of you? Their progression is easier on the other path. None of these are logic failures. They are design failures, the same ones a half-finished world teaches you to see.

Most commercial thinking quietly assumes people are rational. Give them the better product, the clearer data, the stronger argument, and they will act accordingly. Worlds teach you that this is not how people work, and that pretending otherwise is how good plans fail. People act on what is rewarded, what is easy, and what they can see working. If you want different behavior, you do not argue harder. You change the conditions.

Designing for behavior, not for logic

When I design a commercial system now, I think the way I think about a world. What behavior do I actually need. What incentive is producing the behavior I am getting instead. Where is the feedback too slow for anyone to learn from it. What does progress feel like for the person I need to move, whether that is a rep, a partner, or a buyer. The questions are the same. Only the stakes are different.

This is not about manipulating anyone. A well-designed world is not a trick. It is a place where the path you want someone to take is also the one that makes sense for them. The same holds commercially. The goal is alignment, not pressure: build the system so the behavior you need and the behavior that serves the customer are the same behavior. When those two diverge, no amount of messaging saves you.

I did not set out to learn any of this. I set out to build something fun with nothing riding on it. But the longer I spend designing worlds, the clearer it becomes that I have been studying one thing the whole time: how people behave inside the systems we build for them. The domains could not look more different. The questions underneath them are the same.