If you've been following the development of SpyParty, you may be familiar with its lead developer Chris Hecker. A veteran of EA and Microsoft, he's been working on SpyParty since 2009, and in that time has discovered new insights both in game development and the business of game development, many of which he's been able to turn around and use while making SpyParty. 

Today, over on the Gamasutra Twitch channel, we were lucky enough to be joined by Hecker to discuss his experience making the game, and explore his game design process now that the game has been released on Steam's Early Access platform. 

As always, the full conversation is available for you to view up above, but in case you're listening for "banana bread" as we speak, here's a few quick takeaways from our conversation. 

Listen to your game as you develop it

One of Hecker's biggest game design lessons was a notion explained to him by developer Jonathan Blow, which is the concept of "listening to your game, not just your players," as you're developing it. In Hecker's own words, SpyParty was conceived as a game that would resemble Go, with the rules being fixed and there being perfect strategies that players would test, when in reality it's turned into something closer to Poker, complete with bluffs and strategies that respond to an evolving game state. 

The thing is, SpyParty didn't have to go down the Poker route, and Hecker could have continued to make that Go-inspired version of the game, but as the community grew and Hecker observed what they were doing with his game, he explains that following that path of development became more imperative because it built more soundly on the game's foundations, rather than the version of the game originally in his head. 

It's good for there to be multiple pro playstyles

The SpyParty community has grown to the point that there are now "professional" tier players, who, in Hecker's own words, are essentially playing a different game than lower-tier players. In our chat, Hecker outlined 3 playstyles---one that's an advanced version of 'early game' strategies, one that relies on facts about player behavior and human psychology, and one that closely analyzes the behaviors and animations of NPCs and instantly acts on anything outside those normal bounds. 

Each of those players has different needs, requests, and response to fixes, but Hecker said having them is important for the game's long-term health, and it's better for there to be 'no one right' way to play. He said their presence helps him understand what changes are needed for SpyParty's design in the long run, and when navigating feedback from those players, his ability to understand what each of the three "hydras" (quick reference to the Odyssey) need helps him negotiate how all that free player feedback can be translated into solid design.

Use sales numbers to create possibilities, not guarantees

Early on in our chat, Hecker lamented the recent death of SteamSpy, which served an approximate role in helping developers determine how well their own games might perform when released on Steam. After a quick disclaimer that without full transparency from developers, there's no real way to determine how many copies a game has sold and in what ways, Hecker explained that studying SteamSpy helped him come up with a process that he hopes other developers can emulate if people like him continue to release sales data (something he says he's working on).

Per Hecker, he was able to compare different games he studied on SteamSpy not to create precise estimate of how well the game would do on Early Access, but rather make a range of sales numbers he could use to determine how development would proceed afterward. 

In order, that range included "PlayerUnknown's Battlegrounds-level success," "well enough to hire a team," "let go of his artist but continue sole development," and "abandon all hope and return to triple-A."

With the 1st and 4th outcomes being somewhat outliers in nature, Hecker was able to explain that since launching on Steam he's been able to actually pay himself for the first time in years, and give his artist a raise (he clarified he wasn't necessarily planning on hiring a team for SpyParty, he was just using that as a benchmark), putting him somewhere in the middle of his expectations. 

But no matter which of the four possibilities came to be, Hecker's use of SteamSpy not to inflate outcomes, but create a range of possibilities is a model other developers can learn from (if they struggle to figure out how to research data with the site's closure).

For more developer interviews, editor roundtables, and gameplay commentary, be sure to follow the Gamasutra Twitch channel.