Killing Our Darlings: Creating a Game Through Failures and Feedback
Common advice to creatives is to “kill your darlings.” This game, as it currently stands, is barely recognizable compared to its earliest versions. Many ideas didn’t survive the brutal culling of our personal darlings.
At this stage, we have functional gameplay, a matchmaking lobby, deck-building capabilities, and a framework for testing our player-driven narrative. Now, it’s time for another playtest and a fresh wave of player feedback.
In the beginning, our prototypes were on paper. We would deface any trading card we had on hand—commons from dead and active games perished by the dozen in those early days.
Our goal was simple: to make something that resembled a game.
Do the rules make sense? Does the framework… work?
Instead of destroying our entire card collection, we soon started using index cards backed with regular cards and protected by plastic sleeves. Our first real test came at a large gathering of friends and family (the nerdier ones), who played the game over a day or two. Everyone seemed to enjoy themselves and, more importantly, generally understood the gameplay.
Some “feedback” moments were eye-opening, and this playtest offered one of the most memorable.
After clarifying the combat system to a friend who was explaining it to new players, he said, “Oh, I’ve been teaching it wrong all day. How’s that for feedback?”
Clearly, combat was too complicated.
The next iteration moved to Tabletop Simulator (TTS). Although we didn’t add automations, TTS sped up gameplay significantly. We didn’t hold any large playtests on TTS, focusing instead on many test games with a smaller player pool. TTS is an incredible tool, but it requires every tester to have the software and know how it works. Around this time, we realized that the game was taking too long. A 45- to 60-minute strategy game is fine for two friends at the same table, but it starts to feel excessive online.
Based on one player’s comment, “it takes too long for my Agents to get to the missions”, we simplified and condensed the basic turn structure to speed things up.
Eventually, we started active development on a digital version. Once we had a working prototype, we gathered another group for playtesting.
It was a minor disaster (perhaps a topic for another time).
Although the turn flow and cards were functional, the game crashed about half the time, which made it hard to gather reliable feedback. Despite the technical setbacks, we did receive some valuable input…
Nobody liked our economy system.
Comments like “So I spend all this time creating a deck, and the first thing I have to do is discard two cards?” and “I see what you’re going for, but you either need to cut it or lean into it” made us see that this was a painful mechanic for the players.
After some reflection, we implemented a much less frustrating system, and, as suggested, we leaned into the concept of resource scarcity, transforming it into “resource management,” a central challenge of the game rather than a frustrating limitation. By reframing how players manage their decks and introducing new mechanics that reward strategic thinking, we turned what was initially a negative experience into an engaging feature.
Through each playtest and round of feedback, this game has evolved beyond what we could have initially imagined. Each iteration brings new insights and changes that make it stronger and more engaging.
The feedback we receive isn’t just about mechanics; it shapes the experience we’re trying to create. As we move into the next stages of development, we are excited (and maybe a little nervous) to see what surprises the next round of player feedback will bring…
And we’d love for you to be a part of it. Sign up here to join our other playtesters and help shape the future of Agency.
