The sales & design thinking that gave SpyParty a strong launch

SpyParty, which has been in development for nearly a decade, launched earlier this year on Steam Early Access to strong sales and with a large community ready to welcome players to 1v1 matches of psychological warfare.

Booting up the game will give developers plenty of opportunities to study its underlying game design, tutorialization, and matchmaking, but what other lessons can developers learn from the success of developer Chris Hecker's carefully crafted passion project?

Luckily, you don't have to work on a game for 10 years or be making a 1v1 multiplayer game to pick up what Hecker is putting down. We recently chatted with Hecker on the Gamasutra Twitch channel, and we did our best to try and understand what he's learned about game marketing and game design while releasing SpyParty to the public. 

Below, we've excerpted two key parts of our discussion for you to read, one focusing on the sales plans Hecker made before pushing the big 'launch' button on Steam, and the other analyzing the different advanced playstyles of SpyParty, to help you think about how your players might interact with your multiplayer game. 

Bryant Francis, Editor at Gamasutra
Alex Wawro, Editor at Gamasutra
Chris Hecker, Indie developer of SpyParty

Let's talk sales numbers for an indie in 2018

Francis: Your development process is pretty cool, but it's also interesting because you have a background in AAA at large companies, which are notoriously fickle about what numbers they release to the public and to other developers. As an indie developer, what have been your thoughts about sharing this kind of information, and how do you think, looking at when you started, how other developers can take advantage of this kind of information when they can get access to it? Because it's one thing when one can get something like SteamSpy or the other numbers gurus out there. We have a post on Gamasutra, on Steam wishlists.

Wawro: On the front page of Gamasutra is a very good blog about how important Steam wishlists are.

Hecker: There's wishlist numbers. People are also trying to figure out exactly what the multiplier for reviews to sales is, and trying to bound that kind of thing. I'm personally in favor of transparency. I have some developer friends who are not as in favor. I am much more in favor of it than not. I have to find out from Steam what I can share, as there are certain NDAs stuff involved, but I plan on sharing as much as I possibly can, just because I think it helps people make decisions.

I think things like SteamSpy were a net positive. There's certainly some negative things about it. If you just take the number and multiply it by the sale price, that's not what the people were making, because there's sales and such. It's easy to be misled by those numbers. But I think if you were intelligent and understanding of how the business works, those numbers only helped.

They allowed you to go, for me for example, it definitely helped. On the run-up to Steam, I had a ton of anxiety over whether the game was going to sell. I invested a ton, and I'm very proud of the game. Even if the game sells zero copies on Steam I'm super proud of how it's coming out.

Clearly, it's a huge investment both creatively and financially for me, and so I was super nervous. Running out of money and borrowing money from your mom when you're forty-seven and have a child is (laughs) maybe not the best place to be emotionally. But I was really confident about how good the game was. And I can program a computer, so I'm not going to starve in the Bay Area.

I had these four potential outcomes that might happen when it went on Steam Early Access. Obviously, one was PlayerUnknown's Battlegrounds, which was pretty unlikely: mega blowup insanity. One down from that would be, sold incredibly well to the point where you could easily hire a team if you wanted to, but not PUBG's 22 million copies in a year, which was an insane outlier. Using PUBG is kind of like using Minecraft as an example. Let's put the outlier way out there.

There's like really solid, clearly a hit from the get-go. Then down from that is, it's selling. Maybe the burn rate pays for me, but maybe I have to lay John (Cimino) off. There's that middle ground area.

Then there's where it sells zero copies. Somehow all the people who bought it on spyparty.com are the only people who ever wanted to buy the game. I'm right in the middle of those, which is great. But there was this anxiety about that. One of the things the numbers helped with was seeing, in-between the number one and number two thing, between touch-and-go but pays for the burn rate for at least just me, versus being able to hire a team, not that I would, in terms of finances.

Being right in-between there, I took a lot of solace in the fact that there have been plenty of games that have not done as great on Early Access launch, than by doing the Valve-style, big-name updates type thing, but have ended up selling well in the long run. For example, Subnautica is kind of my go-to for this. It came out, they had a pretty quiet launch, and then eventually built it up to selling two million copies before they had their 1.0, out-of-early-access release.

Slay the Spire actually, people don't know this, but that game launched in November and it didn't blow up until January or February. Rainbow Six Siege is kind of the triple-A version of this that everyone talks about. It had a really slow launch, but built it up with the updates thing. There's this talk on YouTube, from Steam Dev Days years ago, by Robin Walker about the big named update thing, about how on Team Fortress 2, they were doing weekly updates and maintaining, but not growing, their player base. Then they figured out that doing these big themed, named updates is the way to do it, and he shows numbers and all that stuff for that.

And so I could have the faith that that was possible, assuming it didn't sell zero copies and I had to get a job immediately. The fact that that's possible is great. The only thing that makes it possible to know that is that people are sharing numbers. People are sharing the stories of what happened. So I'm a big fan of that. The movie industry shares all their numbers, and I think that's great.

Francis: They share certain numbers.

Hecker: They share more numbers than we do, let's put it that way. I think the more, the better, because it helps people make decisions. I think that concrete example, of me going, "Okay, things are going to be okay if I don't have the amazing launch, it doesn't have to sell a hundred thousand copies the first day for it to be successful in the long run."

Designing for different playstyles

Francis: How do you think the different sniper styles have evolved? That question comes from elvisnake. I actually try not to look at who I think is a suspect. How do you design for giving snipers the ability to mess with their prey?

Hecker: Elvisnake is a pretty experienced player. There's the "elite level" sniper playing styles, then there's tools for all levels to do things. The elite level question is really about... there's not exactly a clear breakdown, but there's three kinds of snipers at the elite level right now. There's campers, behavior-less, or behaviorist, there's a fight in the community over what to call them, and then there's etiquette snipers. I'll just describe what those are really quick. Most normal people are a combination of all of those. We have a few savants that are really far at the extremes of these kinds of things.

A camping sniper is like the elite level of the beginner statue camper, except, in order to camp an enemy on four out of eight map layouts, you have to camp all of the hard tells. You don't know which ones they're going to do, so you're constantly task-switching between them. Kalie, who's one of our top players, who's our preeminent camping sniper, goes between: look at the guest lists, look at the statues, look at the ambassador, and repeat. Constantly switching back and forth between these things, while trying to, if bananabread goes off, getting lowlights for that, watching the briefcase. Has to watch all the hard tells in kind of this CPU task-switching, constant kind of checking thing. So that's camping.

Behaviorists are the people who look for people acting funny. Things that are suspicious, that indicate that a person has intent. None of the party-goers have intent except for the spy. They're all just hanging out having a good time. Whereas the spy's gotta move between missions or they're going to run out of time. So, looking for things like intent. The canonical example I love to give is, at the far end, Kalie's the preeminent camping sniper, and has discovered this thing. There's a million of these things, but the one that's easiest to communicate fast is, Kalie's figured out that spies like to talk second when they enter a conversation. Talking first, walking into a conversation would be suspicious, right? So I'll wait for somebody else to talk first, and then I'll talk after that. So Kalie started highlighting people for talking second, which makes no sense whatsoever, but Kalie has shot a lot of spies doing it. That's a pure behaviorist strategy.

Then there's etiquette snipers, who are really into the algorithms the game uses: pathing behaviors, when animation breaks happen, thing like that. So they're watching for things like, did you leave that conversation before your second idle animation was done? I'm going to shoot you for that, or highlight you at least for it.

Francis: You just broke my head, Chris.

Hecker: Yeah yeah, all different kinds of really hardcore game... they're not bugs. Things that are really hard tells or anti-tells. A tell bug is something that implicates a spy, an anti-tell is something allows you to do a hard lowlight on someone, meaning that only NPCs do this. Those are priority-one bugs, we fix those as fast as we can because those break the games.

The etiquette stuff is a little more subtle. Etiquette stuff isn't really game breaking, but I feel like etiquette snipers right now need a little bit of a nerf. I want to make more outros for animations and I'm going to replace the pathing system so they're going to have to relearn everything. Which is great. They love doing that kind of thing. They say that they hate it, but that's what they live for.

So those are the three main, elite-level sniper styles. And then at the lower level, like you were talking about, there's a whole bunch of ways. The technical term is baiting, with the laser. Especially on a map like Library, or even on Courtyard, which is the one where you spin around. Basically you aim your laser somewhere else, but you're looking at the side of the screen hoping the spy does missions. And there's various different techniques for doing that, catching bugs and things like that. When I say "bugs," I mean bugs in the [indistinguishable – eyeglass?] It's a little bit of an unfortunate term.

So there's a ton of stuff like that. Highlight/lowlight strategies. There's things like resting your laser. Laser burning is a thing, where you rest your laser on someone. NPCs can't see the laser. The spy has infrared contact lenses, so they're the only one that can see the laser. The dog can see it too. The dog in the purse can see the laser. But those are the only two people that can see the laser.

Wawro: People? Characters?

Hecker: Resting your laser on somebody to prevent them from doing missions is a thing.

There's a whole bunch of different things like that. We provide affordances for the sniper to mess with the party. And then there's gonna be, as that dossier thing, the third feature in that dossier set of features, the game design features that I didn't mention, the most experimental one. It's called interrogation. The sniper's going to be able to tell the security guard to go interrogate someone. He'll pull them aside, out of a conversation, and there will be some kind of branching conversation thing where the security guard will ask them about something going on at the party, like "What did you think of the statues?" and you'll need to think on your feet. "Have I been to the statues yet? Should I try and bluff this guy?" And the sniper can either zoom in and listen in on the conversation, or can just get a report back from the security guard. There will be more ways for the sniper to mess with the party.

This conversation has been extracted from a longer livestream with Chris Hecker, which you can watch in its entirety right here. Be sure to follow the Gamasutra Twitch channel for more developer interviews, editor roundtables and gameplay commentary!