Speaking exclusively to mobilegamer.biz, Supercell CEO Ilkka Paananen and former Hay Day lead Stephan Demirdjian reveal Team Lab, the new way game teams get greenlit at the Finnish firm.
Demirdjian is now heading up the new process, and it sounds intense. It involves team psychologists and stress tests of the potential cells’ ability to work together and overcome adversity.
Perhaps most interestingly, Paananen and Demirdjian are also now welcoming pitches from external teams who think they have what it takes to get through the Team Lab process. If successful, they can go on to make the next Supercell game.
Paananen also gives us his thoughts on the hurdles games need to clear before they go global, why you need to be a “little bit crazy” to start a new game team and if Supercell could soon expand beyond mobile.
The conversation below has been edited for clarity and length. Readers should note that while the greenlight process is referred to as Team Lab below, the programme will be known as Spark going forwards.
Can you explain to us what Team Lab is and how it has changed your processes?
Paananen: So basically it started last year, we started to think about how to make Supercell an even better environment to create the best new games. And then Stephan and a few others made it almost like a research project.
We spoke to all kinds of people in games, but also outside games, we talked to investors, founders, accelerators, incubators, even some members of academia – all kinds of people.
And when the answers came back, there were two thoughts that I feel have always been at the core of Supercell: it’s all about the team. A fantastic team is the starting point, and that should be the singular focus.
And then the second one is that somehow you need to create an environment around the team where a team can have the biggest possible impact.
And then of course, third – you also need to get a bit lucky, because obviously there are lots of great teams out there that work super hard and are trying to create something new, but it just doesn’t work out. You’re actually more likely to fail than succeed.
So you actually want to have people in the team who actually enjoy this type of thing – you have to be a little bit crazy, I guess, to start to do something that you have to work super hard on, and yet the odds aren’t really on your side. But you have to somehow get this energy, and you have to thrive in that type of environment.
Then we started thinking okay, what does it mean in practice? How do we actually make Supercell better in those two things, and not just better, but ultimately, the best?
And then our thought was to put together Team Lab, which is all about putting together the best possible teams, rather than these teams spinning up organically or on an ad hoc basis. We wanted to get super systematic about this and consolidate and accumulate the knowledge we have about building teams. We have 13 years of history, so let’s make sure we learn from the 13 years.
We set ourselves a goal: we want to be the world’s best at creating these new game teams. I truly believe that one of the core pillars of Supercell culture is this idea that we don’t greenlight ideas, we greenlight teams.
It’s a kind of a cliché, but a great team can make an average idea great. But an average team with a great idea most likely won’t succeed. The more pragmatic point is that when a team starts to build a game, if any game comes out, it’s going to be different than the original idea. So if that’s the case, why would you even bother thinking about the idea, or approving or realigning the idea?
We’re going to focus all of our attention on the team, and put in place an amazing methodology including even having a professional psychologist in the team assessing the team dynamic.
So this is a way to structure the formation of new teams and games more effectively? Can you tell us what’s involved in that process?
Paananen: That’s exactly where we wanted to see an improvement, we weren’t systematic enough. Everybody has a game idea and lots of people want to create new games. But if the goal is to build the best possible teams, how do they get there?
And by the way, there’s very little process and structure and Supercell – we tend to be quite allergic to it – but this is one place where we felt that we absolutely needed it.
Demirdjian: Since the company is growing now, I felt like we needed to formalise the system behind how new game teams form at Supercell for our new employees, but also the older employees. I’m not sure if we ever had a really clear cut answer for everyone.
We built two core pillars. One was to formalise greenlighting teams, not the ideas. We assembled a greenlight panel that interviews those teams and the chosen leads, because it was important to me that it shouldn’t be one person that is the gatekeeper. It’s more like: if you guys believe in this, and you can convince me, then by all means, go to the greenlight panel and hopefully, you will pass.
Who’s on the greenlight panel?
Demirdjian: For the first run we selected a really diverse panel and Ilkka was part of it. I was very proud of having experts that we really trusted from all parts of the company. It’s not always the same people that make those decisions, or that you have to convince in those interviews. The aim with the second round is to definitely infuse additional people into the panel.
So what is that panel is looking for, and what criteria are teams greenlit on?
Demirdjian: We want to infuse, from a cultural point of view, more entrepreneurial thinking. You have to have the right ambition levels, you really aim high, you don’t let yourself be scared or think too small.
Of course we need to also have people that are able to kill their darlings, that are able to make those really difficult decisions – killing your entire game, maybe even having to kill your entire team, even.
You have to also be hungry and you have to be able to work with constraints, so it really takes a specific type of developer. We basically test different team compositions, and together with our performance coaches and psychiatrists, we evaluate if the team going the right direction, can they work in different styles, can they create a vision together?
I have personal failures on that front, when my past new game teams fell apart. Sometimes it’s really those small interpersonal connections that create the biggest problems. My hope is that we frontload all those learnings from the past on why teams failed. Hopefully this way we can strengthen the creative core of a team and say: this team is really strong enough because they’ve gone through shit together.
Through that stress test, they have also really deeply collaborated with each other, they know each other well, they know also how to handle conflict and all the stuff that goes into efficient teams and high performance teams.
Paananen: The core team has to feel like a team of entrepreneurs, as if they are about to build a new company.
Do the teams have to come to you with a game prototype and a team vision? Or can it just be a team alone?
Demirdjian: We have completed one test run and yes, as part of the deliverables and the pressure, they have to work on prototypes. But it was important for me that the Team Lab lead should not have a say in what kind of games they’re building in those test runs.
I wanted to approach those prototypes more like a tool to see if this team actually can work together and can create something they feel proud of. But at the same time, can they also deliver based on the deadlines they’ve set up themselves, together with me?
If the team feels they are ready and every team member can tell me ‘I’m willing to fight for every single person in this cell because they make me better’ – and I agree – I put them into the final phase of the programme. That means getting ready for greenlight, and they can work on the next Supercell game.
Does all this suggest that you’ve gone too far into development with some games in the past, and you need to pull back a little? And how does that balance out with the need to release a new game? It’s been five years since Brawl Stars…
Paananen: It goes without saying that the bar for releasing new games is different to what it was six years ago. Players’ expectations are higher, and the market is even more competitive. That’s just a fact. We just need to have even better teams.
I would almost say that we are doubling down on what actually makes Supercell Supercell.
Previously these cells were organically formed and were a little bit more ad hoc, and now we actually want to put attention and focus in even forming the cell, because we know a once the cell is formed that cell will go their own way. And then they can obviously kill the game, or release it or do whatever they want with the game.
What hurdles do your games need to clear to go global? What is the criteria, is it financial, retention, player feedback…?
Paananen: Nothing has really changed from that point of view, our bar continues to be super high. Day 30 retention and long term retention are most important metrics for us.
Of course, we do care about monetization, although we do believe that you can build monetization later on, because if you manage to build something that lots of people love and are super engaged with, we believe that over a long period of time you’ll figure out how to how to monetize that experience. But you need to create something that is so fun that people come back time and time and time again.
Things like organic installs is an important one. We aren’t that dependent on user acquisition – over 80% of our new players come from organic sources and I think it’s probably exactly the opposite for quite a few companies.
Our teams have been able to build these great games that are worth talking about, so therefore they spread virally. That’s an important factor in evaluation as well.
Have you seen any results from this new process yet?
Demirdjian: In a test run that concluded just now in February we actually did greenlight two teams, and we also killed a lot during the Team Lab phase.
We’re quite happy with the results with this initial test and as a result, the decision was made to do another round, which will start after the summer. And this time around, after speaking with Ilkka about this, we want to also open the opportunity to external teams to be part of the journey.
Paananen: There are alot of great people, and great ideas also outside Supercell and we want to make sure we’re tapping into the world’s best talent. We want to welcome people who will come in and even challenge how we do things and what type of games to build.
We’d love to have people who are thinking about games differently and would build something that maybe we wouldn’t build. What we can offer those people is this environment where if you truly want to work with the world’s best, you can be part of it.
We have a very, very thoughtful approach on how we put together these teams. It’s not as simple as putting together the world’s best artists, designer, client engineers and server engineers and magically it will work.
The psychologists we have work with professional sports teams at the highest level. You would be surprised by the similarities in putting together a professional sports team and pulling together a new game.
They spend so much time assessing not just the hard skills, but but the soft skills, and how these people fit together. During the first Team Lab run we killed five teams, and then two teams passed – that gives you also an idea what we’re doing here. And we’re excited about the opportunity to open it up externally.
Do those pitches have to be mobile games or mobile-first games?
Paananen: We are starting now mobile first, you know, we’ll see what the future holds. But I’m sure you heard about our North American studio – especially PC will be a bigger and bigger thing. But we’re still a mobile first company and we wanted to stay true to that, at least in this first phase. It might change going forward.
