How Canada’s Behaviour Interactive hatched its horror hit Dead by Daylight

Dead by Daylight could be the biggest success that Behaviour Interactive has seen in its 24 year history. The Canadian game studio, which spent much of its existence making games for other game developers and publishers as an outsourcing contractor, made the horror game with publisher Starbreeze, and it became big hit upon its debut in June. That could help it become an original game maker in the future.

The title has shipped well in excess of a million units, and it is growing more popular with each month. And Behaviour and Starbreeze launched an update for Halloween — with addition of killer Michael Myers of the Halloween movies. That led to a record month for engagement in the month of October, said David Osborne, vice president of product at Behaviour Interactive, in an interview with GamesBeat at the company’s Montreal headquarters. I also played a couple of rounds of the multiplayer game, where you play one of four human-played victims trying to escape from one human-played killer. It was both terrifying and fun.

[aditude-amp id="flyingcarpet" targeting='{"env":"staging","page_type":"article","post_id":2112252,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"business,games,pc-gaming,","session":"A"}']

“Our company has reached a new level of success with Dead by Daylight,” said Remi Racine, CEO of Behaviour, in an interview.

During the Montreal International Game Summit (MIGS 2016) event, I visited Behaviour and caught up with the executive team. That included a joint interview with Osborne and Stephen Mulrooney, chief technology officer and vice president of Behaviour Digital.

They talked about their Hugo research and development division and how it gave birth to Dead by Daylight, which won a couple of trophies at the Canadian game awards. Behaviour has split itself into multiple divisions, including parts that do traditional game outsourcing and others doing original work.

Here’s an edited transcript of our interview.

GamesBeat: You’ve had great success with Dead by Daylight. But tell us about Behaviour Interactive’s history first.

David Osborne: This is the stuff we’re prototyping. It’s very rough. There’s no art, as you can see. It’s just a proof of concept. We’re looking at a control method that could be good for VR. It’s one problem we’re looking into.

Behaviour is an umbrella. We’ve been around 23, 24 years in Quebec. Traditionally a work-for-hire studio, but we’ve done some original IP over the years, starting with Jersey Devil back in the ‘90s on the PlayStation. We also did Scaler, and then of course Wet and Naughty Bear. We’ve had a cadence, every few years trying to do something ourselves, but always with a publishing partner.

[aditude-amp id="medium1" targeting='{"env":"staging","page_type":"article","post_id":2112252,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"business,games,pc-gaming,","session":"A"}']

We decided a couple of years ago to break off into separate studios that were specialized. Behaviour Digital, which is us, we do our own IP. We also license IP, but the idea is we’re co-publishing with partners and making our own product on our own terms. Behaviour Studios is the traditional work-for-hire studio that’s still going very strong. It’s one of the cores of the studio. Behaviour Mobile specializes in mobile publishing. Then Behaviour Business Solutions offers game design thinking for serious applications like training for business. All four studios are in this studio, with the exception — we also have a Chilean studio, which helps out with Behaviour Studios. They’re down in Santiago. That’s our only satellite studio.

Stephen is head of publishing and CTO of the whole company. I used to be head of creative for the whole company, but as we broke out I’m now head of product for Behaviour Digital. My concentration is just on our products now. Stephen’s been in the industry for more than 20 years, starting at Ubisoft. He’s been here now for 16 years. I’ve been in the industry for 25 years, starting over in Scotland at DMA Design working on Lemmings and Grand Theft Auto. I’ve been at Behaviour for 12 years now. We’ve worked together for a very long time. It’s great now that we get to work directly on our own stuff.

The vision for our studio — independence is our ultimate goal, through creative control, good choices in production and business, and player engagement. Obviously without that we go nowhere. Our mission is to make remarkable games that we would play, our way. Not only making them our way, but also playing them our way. We believe in player choice.

GamesBeat: And tell us about the research and development you do.

[aditude-amp id="medium2" targeting='{"env":"staging","page_type":"article","post_id":2112252,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"business,games,pc-gaming,","session":"A"}']

Hugo is the R&D group that I formed a few years ago. It’s a small group of people, people coming in and out. There’s a full-time creative director and technical director. An art director is in there helping out right now. In the past we’ve had full-time QA when it’s made sense, because we’ve had builds we wanted to show and get some feedback. We’ll do a build and then QA would take it and do a focus test internally. Then, at the next step, we’ll start showing it externally.

Hugo is just the name of the group. It’s not a product as such. Historically we’ve had two teams. We have three now. Hugo would run in parallel and feed the current games in production with ideas. The games in production would also feed Hugo with big expensive things they’d already done that we could capitalize on and reuse. It’s very much all three of them working in tandem.

The market for Hugo is the one we’re setting out to build for studio as a whole. It begins with these two games, Warhammer and Dead by Daylight. Dead by Daylight, we’ve partnered with Starbreeze, because they have the Payday community. For Eternal Crusade obviously it’s a Warhammer game, so we partnered with Games Workshop on the licensing to bring in that market. To build our own market for our studio that we want to grow, we’re doing online multiplayer, mature, PC and console.

We identified survival as something we wanted to look at early on with regards to the kind of games we like. We saw it as underserved, but very popular. We’re exploring it with a lot of our Hugo projects. A Hugo prototype is remarkable. If it’s been done before, the thinking goes, why do it again? The search for the purple cow, that reason to be — in this day and age, with everybody able to self-publish, if we can’t stand out there’s no point. So we start with that, what’s remarkable.

[aditude-amp id="medium3" targeting='{"env":"staging","page_type":"article","post_id":2112252,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"business,games,pc-gaming,","session":"A"}']

GamesBeat: How do you know when you have something?

Osborne: It has to be fun. That’s the entire point of this, that it’s fun. If that’s going somewhere, it takes the next step to QA. Then it goes to internal focus tests, then external focus tests, then we start talking to potential partners and getting feedback from people we know and trust. It has to prove a compulsion loop. There has to be a reason to keep coming back. What they’re not is tutorials. There’s no onboarding. There’s no polish, no art. The audio is functional. It’s not necessarily stable. It’s not reusable code. It’s very fast, rapid prototyping.

Dead by Daylight was the first that used this process. Hugo was born from the success of this exercise. Dead by Daylight itself—as a company we’ve been working and experimenting. You can look at some of the drawings here. We’ve always been interested in being the bad guy. It’s an underserved thing. I remember even early times on GTA where it was like, “You can’t do that!” There were no games about letting you be bad. Obviously GTA goes to great lengths to prove that you’re not really bad. But there was something interesting there.

We did a number of prototypes that led to things like Naughty Bear, where you have to exact your revenge on these things. We tried lots of AI and things like that. But what we found is that real people making stupid decisions is a lot more entertaining than a machine trying to do that. It’s real, right? That very emotion of watching our game, which is a very important part for us in the process—people even liked watching it. A lot of people say, “I don’t want to play,” but you could see that they were gripped. They had emotions about it. It was like watching a movie.

[aditude-amp id="medium4" targeting='{"env":"staging","page_type":"article","post_id":2112252,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"business,games,pc-gaming,","session":"A"}']

We knew that streaming was getting bigger. We knew that influencers and bloggers were starting to have a massive effect on the industry. People watching games was becoming a credible thing. You get a lot of the same emotions. If you see somebody being bad at a game, it makes you feel smart. If you see somebody being great at it, you feel like you’re learning something. But of course now it’s spread to people who aren’t even gamers themselves.

GamesBeat: How did you manage Dead by Daylight?

Osborne: So we were thinking about this for years. Let’s find the fun, ask others, and then greenlight it. That was the process. We started with a Hugo prototype. It was two staff, two creative directors, and myself experimenting with what this game could be. Because it was a rough prototype — it wasn’t even networked — we simulated the multiplayer experience by doing split screen. We had a piece of cardboard cellotaped down the middle of the television. We’d bring people in to do tests. It was a simulated experience. Instead of having four playable survivors, we just gave you four lives to try to simulate it.

Early on we found out that the monster being first-person, them being third, was a really good way to go. I can show you a video of that, the very first prototype. No sound or anything. You see here the survivor. All these meshes are just pulled off the web. The whole thing was programmed by one guy. We can see the killer with his killer’s eyes. He’s got the red light. He’s first-person. They’re third. Climbing through windows and things like that. He’s looking for them. There he goes.

[aditude-amp id="medium5" targeting='{"env":"staging","page_type":"article","post_id":2112252,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"business,games,pc-gaming,","session":"A"}']

People seeing that, even that, would say, “Augh!” Even with funny art and no real sound effects, we could tell something was there. It was organically fun. Matches were very fast, between six and eight minutes. Every time you played it, it was different. Because there was no AI, there was a feeling that the moments were real. We’re going for great moments, because that’s what Twitches well, but it’s also just what you remember about a game. You remember the high points and you remember the ending.

We’d set this up in a dark room and get people to come in and play and watch. Our greenlight process was just that, watching everybody reacting to both playing and spectating. We’d massage the prototype, add features or take them away. The levels were generated every single time, because we didn’t want anyone to feel comfortable in mastering the maps. We wanted people to feel lost. If one of you wasn’t lost, it would be unfair. The killer doesn’t know the map. The survivors don’t either. You roughly know the map, because of a few landmarks, but other things change every time you play.

That’s very important to us at our studio. Smart choices affect content. Smart content and content creation is a big pillar for us. We’ve made a number of titles where every time you play, it’s different. The next stuff we work on, we take that further.

GamesBeat: This sounds like the opposite of Evolve.

Stephen Mulrooney: To come back to the point about different perspectives in the game, with the survivors in third-person and the killer in first, that was proven by the prototype. Yet, it was so counter-intuitive for everyone. If you asked anyone before they played the game what the correct way to do it would be, they’d say the killer should be powerful, so he’s in third, and the survivors should be immersed, so they’re in first.

When we inverted it, though, the flip side was true. The great moments are when the killer walks by you while you’re crouched in the grass and his constrained field of view means he doesn’t see you. You manage to survive because you were just in the right place at the right time. It was something that only came out because we did the prototype first, because we found the fun first. That’s what we’ve kept doing since.

Osborne: We knew we had something special on our hands early on, and that’s what Hugo is for. This would be the point where we’re saying, “It’s okay, but it might not have legs,” and that’s when we’d just move on to the next thing. But this one, it was very much working.

The next step after that was to go into preproduction. We increased to 10 staff. We needed to network it, first thing, so we could try four on one, five on one, six on one, two on one, find out the right balance and the right map size. That’s also when art started getting involved, thinking about the tile system and so on. Still very gray-box, but with proper metrics and so on. It’s not VR, but we did do tests early on in production.

GamesBeat: What did you settle upon for the final game?

Osborne: The incredible thing is that nothing’s really changed, and that was two years ago. The reason is that it was all pre-validated with the prototype. When we ramped that up to 10 people, this stuff wasn’t throwaway. We were doing proper code, building it out properly. Then we go into full production, ramping up the team. At the end of preproduction we had a fully playable gray box experience, four players, and that’s when we did the green light.

The offering at launch had three different killers, four survivors. We’re starting to layer on top of that with DLC. We had a ranked mode first thing. Then we added Kill Your Friends mode. We have three different locations, 12 maps, and we’re adding new maps, new killers, and new survivors. Eventually we’ll add new modes like competition. The blood web, the way you do your progression, is quite unique to our game and unique to each player. Our first DLC was free for everyone as a thanks, because we had a very successful launch. We came out at number one on Steam. We were number one on Twitch that week. It was our best E3 ever. We couldn’t be happier.

GamesBeat: And what’s the result?

Osborne: In July we added a lot of new things like new languages. We’re adding Chinese and Russian. We’ve added new maps, survivor outfits, and key bindings. The plan is to keep the product alive with the live team now. In October we had a big release where we sub-licensed the Halloween IP and put Michael Myers in our game. The timing was brilliant. The sales were really good. The reviews were fantastic.

Total Twitch time has been 200 years a month. That’s how much people watch. Three and a half million hours watched. We do our own Twitch every week as a team. We talk directly with Twitch about things they’d like to see in the game. They’ve been a big vehicle for us, and for future games too. We’re looking at what makes great moments, why something would be interesting to watch.

Average play time is 35 hours, which is crazy. Especially when you look at how quick a game is, six or seven minutes. But that’s part of it. “Oh, I’ll just do one more.” A lot of people play both killer and survivor. At the beginning we wondered whether it would end up being one or the other, but there’s a big crossover. People like to do both. It really is like playing two different games. They have separate progressions as well.

GamesBeat: Where is it popular?

Osborne: Demographics-wise, we’ve been really successful in Asia. We’re very happy with our success in China and Thailand. We’re looking to add other Asian languages. We have a very big percentage of women playing.

Back to Hugo, in the last year and a half we’ve done 10 playable prototypes after Dead by Daylight. Each of them have their own legitimate future. None of them are cancelled per se. I’m very keen to often drop something and come back to it later, or to be working on multiple things in tandem. We’re never just obsessing on one thing. I want to keep everyone fresh. We can have an idea and jump back. “That would be great for Charlie, let’s go back to that.” They’re just named after the radio alphabet, which I don’t regret, because otherwise I could never remember them all.

The idea is that this will go on forever. We’re up to J, and K is next. Once we go through the alphabet we’ll go back around again. I’ve run R&D groups in the past at other studios. It’s a super-important part, keeping that eye to the future. It’s hard to do, especially when you have to put aside some of your best guys, guys who’d be amazing in production. But we have to think to the future. We want a good solid future for our studio and everyone who works here.

GamesBeat: How long did Dead by Daylight take in the end?

Osborne: The actual prototype was about two months. Preproduction was about six. The product itself was a year and a half, two years. Probably two and a half years in total.

Mulrooney: Not counting all the earlier iterations.

Osborne: If you look around here, even earlier ideas for games where you run around killing people — it started as a board game. We didn’t even go near code. We were playing with dice and little guys, lots of hand-drawn art. Again, super rapid prototyping. It’s about getting into the game as fast as possible. How can I show it to people? How can you experience what I’m thinking?

Like any good skunkworks we keep the team sealed off so they can experiment. It’s a dream job, an opportunity for them. Like I say, it’s a very important part of our future as a studio.

GamesBeat: Do you ever bring the work-for-hire clients in to get their feedback on some of these?

Osborne: Absolutely, trusted partners. Starbreeze, we were working with them in the work-for-hire space when we decided to partner with them on Dead by Daylight. We already had a relationship through our work for hire. Also, of course, the legacy of that and our reputation as a work-for-hire studio has been a strong foundation for us to break out into doing our own stuff.

We’ve built connections not just in the business world, but also in licensing. We have a reputation for getting the job done. What we needed to prove as a studio was that we could come up with something ourselves and do it from scratch, our own creative vision. We’re happy with the result.

GamesBeat: Does a client give you some of that needed feedback on Hugo projects? Or do you guys mainly use the company itself to do that?

Osborne: The company itself. Normally it’s progressed beyond Hugo by the time we show it to a potential partner. We don’t show everybody what you’ve just seen. Some of that could come to fruition years from now, as an idea or a concept. I do have a tendency to keep my cards too close to the chest. But when it makes sense — if you look at Dead by Daylight, once we got to the end of that prototype and we were starting preproduction, we found that the more eyes on it, the better. The more criticism, the better. The greater the input, the better. There’s a point at which you need a lot of feedback, especially when you’re deciding whether or not to do it.

Mulrooney: There were people we showed Dead by Daylight to who we know we were never going to work with. It was just to see them play the game, to get their opinion, and to see them watching the game. That was very informative for us as far as the decisions we made and the direction we went.

Osborne: Every company out there has specialist people who do this. In my previous position, I was at Jaleco, Japanese Leisure Company, in New York. I was head of external development. I used to be on the other side of the fence, evaluating and looking at hundreds of prototypes. Before that I was in development running R&D. I’m kind of back on the side I started at.

GamesBeat: Where does a storyteller come in? Do you have a backstory that you’ve built for Dead by Daylight or any of the Hugo prototypes?

Osborne: For Hugo it really depends, because at that stage we’re concentrating on the moment-to-moment gameplay. The fiction — there isn’t anything that’s especially story-centric. It’s very expensive to do a linear story-based thing that you’ll go down one way. If you want it to branch it gets even more expensive. A lot of our Hugo prototypes rely on fun multiplayer interaction.

The fiction for Dead by Daylight we keep deliberately vague. When they were making Lost they said, “We’ll never tell you anything and you’ll keep watching. If you understand things you’ll stop.” I feel like that’s a good way to go. You feed the community with a bit of information and then some of the hardcore will explode. “Oh, is that linked to that?” They can make it much bigger than it actually is. Sometimes you’ll listen to that and maybe even change things slightly because of the community. It’s growing more organically with the product and the DLC as we add new killers.

The entity itself, the darkness that runs this whole eternal nightmare, we don’t explain much about that at all. It’s just there. I feel like the less you explain, the better. The game itself feeds into that. There’s no in-game communication. You can’t talk to each other, unless you’re playing with friends. That makes it very international, too. You can play with people all around the world, because it’s a visual game. We put in some emotes so you can say things like, “Come here.”

It’s funny, though, because the community has turned that into a bitch slap. I was watching one of our streamers the other day. One will go up to the other and do the “come here” like they’re slapping their face. It’s amazing how the game is growing beyond our intentions. That kind of stuff says, “Hey, maybe we should add a slap.” Or if you put somebody on the hook, if they’re being camped because the killer is hiding there, they can use their escape animation, or only part of it, to announce that they’re being camped. We really want to capitalize on that kind of gameplay people have been finding.

GamesBeat: We’ve seen a wave of these kinds of asymmetric games — Left 4 Dead and then Evolve.

Osborne: Left 4 Dead is a little different insofar it’s a bit fairer. It’s against the AI, like Payday. Evolve, of course, we looked at that. We monitored its initial — it sold okay. But it’s just now transitioned to free-to-play. That’ll be interesting to watch. I played a lot of it. It’s quite different there in how you start off so weak. It’s like a scale that at some point goes *donk* on the other side, whereas in Dead by Daylight it’s *donk* right off. [laughs] With the exception of Michael, who gets stronger as you play, it starts out pretty brutal.

Mulrooney: And Michael still starts out stronger than you. In fact, there are people who play Michael to their advantage in the first level of his evolution. It’s easier to sneak up on survivors in the first stage of Michael. You pull them off generators and stuff and just put them on a hook without even injuring them.

Osborne: I always like explaining to people, “As a survivor you can’t damage or kill the killer. You can only escape. And by the way, he’s faster than you.” They say, “How will that be fun? That sounds miserable!” And then you go in and the moments are real. The fear is real. That was very much the pillar of the game – the fear is real.

GamesBeat: Did you ever consider arming the survivors?

Osborne: Oh, yes. They have a torch, for example, which is the nearest thing. They can flash the killer and he’s momentarily stunned. We prototyped guns and things like that. But it’s a balance you really have to watch. Give someone a gun and all of a sudden there’s an expectation of that. Our slashers, or otherworldly creatures, none of them are killable, just like in the movies. Shoot them, bury them, cut them up, whatever, they’ll come back. It’s a fortunate trope in horror.

GamesBeat: What are these drawings in your office?

Osborne: Lots of concepts from other works, prototypes, other horror ideas. We were working on a horror concept at one point where you could possess people and walk around as them. That’s something we’re looking at now for a killer in Dead by Daylight. Some of this work goes back — goodness, this is the late ‘90s, when I was working with Nintendo. Not all of this work is from Behaviour.

GamesBeat: How many platforms is Dead by Daylight on right now?

Mulrooney: It’s only on PC. It could be on other platforms, but it would have to be a different experience, for instance to take advantage of mobile. The IP could definitely end up on mobile someday. But we’d have to figure out how to do that and still keep the core experience in a different way.

Osborne: We took Naughty Bear, which was on console, and put that on mobile. It was our first mobile game. We used assets from console, put them down in Unity, and learned a lot doing that. It was a different type of game entirely, built for touch. But the possibilities for the brand are strong in every direction.

Disclosure: The organizers of MIGS 2016 paid my way to Montreal. Our coverage remains objective.