Supercell: "We don't have a clear structure"
How is the work on games built within Supercell teams, – App2Top.ru told the product-lead Hay Day and Boom Beach.
Ilkka Paananen, CEO of Supercell, previously said that all managerial functions in the company are given to development teams, the so-called “cells” (cell).
But how exactly the processes are going on in the company, whose games earned € 2.1 billion in 2016, Supercell rarely says.
Therefore, this is exactly what we talked about at the 4C: St.Petersburg conference with Timur Haussila, who led the work on two of the company’s leading projects – Hay Day and Boom Beach.
Alexander Semenov, Senior Editor App2Top.ru Questioner: I would like to talk about the Supercell structure. There is not much information about this today. Tell me, how do the “cells” that make up the company arise in general?
Timur Haussila
Timur Haussila, product lead of Supercell: Despite the fact that we believe in “cells”, we do not have a clear structure. We ourselves do not fully understand how it works. But I would say that in many ways it happens by itself. People with common interests, who have time to develop something new, get together and form a “cage”.
As soon as a “cell” is created, it must receive permission [to function] from the CEO. So this is not a model in which everyone does what they want. We are talking about people who are inspired by something, who have an idea, they work on it for a while to make sure it’s worth it, and then ask permission to leave the main project and start working on a prototype.
This is the first test for the cell. A test in which it turns out whether the team can work together and offer something worth developing.
When does this happen? After the release or after the closure of a failed project?
Timur: It can happen at any moment. But this usually happens when the team decides to split up and “kill” the game. At this moment, free employees appear in Supercell. And they are looking for new opportunities. Therefore, this is the best moment to create new cells. At such moments, it is easier to form a cage. Someone can approach someone else and say something like: “Hello! What do you think? Can we work together?” But often employees also leave the operated projects if new opportunities open up before them.
Again, if we talk about big, already released games, then after implementing all the key features, you may well decide for yourself that it’s time to leave the project and take up an idea that you have been discussing with your colleagues for some time.
Is the work in the cells different from each other? Or do they work according to approximately the same scheme?
Timur: Every cell works differently. Even the background of the project managers who are responsible for the work of the cells is different. For example, I have experience as a product manager. Some game leads have programming experience, so they can, in addition to managing the project, write code for it. Some leads came from artists.
Who is our leader depends not on the specialty, but on the team itself.
Does this mean that each cell has its own tools?
Timur: Yes, we use different tools. We develop our own internal engines, but different teams use different variations of engines. Some elements (for example, social ones) can be used by several teams, but in general each team has its own engine. And we don’t have teams that work only on engines, each team builds its own tools.
As I understand it, each cell works in its own way and takes full responsibility for itself. But what does management do within such a structure?
Timur: The role of our management is the role of the CEO. Besides him, we have no managers.
Is there no one else in the department?
Timur: Only the CEO and the game teams. Everything.
The role of the CEO and founder is to help the teams. To ensure that everything works well for them, to help with recruiting, to facilitate the work of teams.
But who decides how much and to whom to pay?
Timur: Project managers together with the CEO and Finance Director.
But you probably have some unique practices?
Timur: We have various bonuses and option programs. We want them to be honest. It’s difficult, but we’re doing everything we can.
Tell me, please, how do you work with documentation? It is generally believed that documentation is not so important for small teams. What do you think about it?
Timur: Of course, this aspect also seriously depends on the specific team. Some prefer to write more documents, but mostly we have very little documentation.
If I have ideas that I want to return to in the future, I usually write a doc for myself, in which I just structure my thoughts. Others do as well.
In the event that I want to share how everything should be implemented, this dock will be a starting point, inspiration. But we do not have these documentation programs, which then work as a reference for the implementation of everything and everything. We have small teams in Supercell. Even in the largest cells, a maximum of three front-end developers work.
So if you have an idea, you just loudly voice it and discuss it right in the workroom with the team.
Now we are working as part of a new game team, there are only 4 developers. And usually it looks like this: “Hey, guys, can we try this?”, “Yes, come on. I don’t think it’s a really cool idea, but let’s try it,” “Hey, it worked fine, let’s keep sawing.”
Something like that.
We have a general idea of what we want to test for playability. And sometimes we have something really working on our hands.
And is it normal for you to do without documentation?
Timur: Yes, of course, it’s fine.
Usually the documentation at the end of the work turns out to be something wrong. It is created as an idea that can be implemented. During development, you change the idea. And when you then look back at the documentation, everything turns out to be different there. This is wrong.
It seems to me that there are two types of work with a project: its development and its support. They are quite different. Do you divide teams into two types, a la, these are responsible for development, and these for support?
Timur: We do not consider them as different types of development.
Game teams after launching a successful game, of course, do not want to abandon the project. They want to understand how it works. They want to operate on him.
But usually it happens like this: you leave, you come back, then you switch to another released project, then you go to work on a new game.
You see, both in development and in operation, you learn a lot. It’s good to work on a new game from time to time, as well as on an existing one on the market. Since, while working on an already operated project, you understand how games are managed in real life, what you need to take on a pencil when developing a new game. If you want to work on a new game, you have a chance to prove yourself from the creative side. So I think both are interesting.
Do you have something like a schedule or an installation, a la “we have to release a game a year”?
Timur: No.
Of course, we have some ideas and plans inside. The teams are working on new games and they have the brightest expectations. But usually we never achieve the planned results. And that’s fine. These are not goals, they are expectations. We have no clear tasks. Of course, we know that it would be great to release one or two games every two years. But this is not a task. If we were faced with such a task, it would be the wrong incentive for the team.
Last question. Today, many teams draw inspiration primarily from the box office top. What inspires you?
TIMUR: Of course, we also look at the box office top to understand what works and what doesn’t. But in general, we are inspired by literally everything. Current games from the box office top, games from the box office top five years ago, games with Commodore 64, Nintendo projects. We draw inspiration from everywhere. Including looking at how people in real life touch objects from walking in the countryside. For example, the inspiration of Hay Day was reality. User interaction [with objects in the game] simulates interaction with real objects. So, yes, we are inspired by everything around.
I see, thanks for the interview.
