The following is an excerpt from the Designer Notes for Offworld Trading Company. The game, an economic RTS set on Mars, releases on April 28, 2016, and is available for purchase here.

I mentioned above that the earliest version of the game had neutral Colonies and no HQs. When we replaced Colonies with HQs, the players were immediately more invested in the game as they felt ownership over a part of the map and could also see the progress reflected in the size of the HQ. However, the game’s fiction suffered some as it was unclear why exactly these companies were on Mars. Were they just manufacturing resources to sell to each other? Obviously, the Offworld Market was part of the answer as one purpose of business was to supply goods to colonists living offworld in the Asteroid Belt. Nonetheless, the game felt like it was missing some focal point to justify all this business.

Thus, we brought back the Colony, this time as a single entity in the middle of the map that would serve as important source of market demand. (When the game had no Colony, the market had a hidden demand curve which drove up prices; putting a visible Colony on the map meant that we could remove this hidden mechanic.) The needs of the Colony were visible in the types of modules which extended out from its center. Habitat modules would need Water, Food, and Oxygen; Office modules would need Power; Laboratory modules would need Chemicals; and so on. Further, the Colony would grow over the course of the game, increasing demand for these resources.

That simple system, however, was quite different from how the single neutral Colony started, and the process to get to there is a good lesson about how game mechanics work best in strategy games when they are simple and transparent. The original system, in contrast, was complex and opaque. (In Sid’s words, the problem with complex, opaque systems is that computer is having the fun, not the player.) Initially, the modules all consumed multiple resources (the Office, for example, consumed Power and Electronics) while also having a cost (the Office required Steel and Glass) which affected the probability of it appearing if those resources cost too much. Further, each workplace module could level up depending on that resource cost as well as the Colony’s excess population that needed jobs. Each Habitat module could level up based on its build costs as well as the Colony having more jobs available than population. Moreover, the Colony’s population would go up and down depending on the price of life support resources, which could then leave some workplace modules unoccupied so that they didn’t consume any resources at all. Thus, the Colony might not be growing because Steel cost too much to construct more modules, or it might not be growing because Food was too expensive, or it might not be growing because there weren’t new jobs available for the new colonists. Got all that?

Most players simply ignored the Colony as it wasn’t clear what was going into the black box or what was coming out of it. The system needed radical simplification. Slowly, each part got stripped away – the resource costs for each module, the population restrictions from life support costs, the leveling up of each module – until the colony itself was simply a visual indicator of demand. Finally, each workplace module consumed only one resource (Offices consume Power, Laboratories consume Chemicals, etc.) and every workplace module had a matching Habitat module that consumed life support. New modules were added randomly over the course of the game, independent of the players or the market. The Colony was simply another random element which added replay value by diversifying resource demand — if the Colony is full of Laboratories, for example, then the players might want to invest in Chemical Refineries because of demand from the Colony.