Designing Two Tasty Cores Three Times Over: The Case of Brawl Stars

Thanks for coming here. It's really nice to see a full room. And uh uh first of all, I'd like to kind of warm the crowd by asking uh a show of hands, how many of you have played uh the game I'm going to be talking about today, Brawl Stars? Nice. Okay. How many of you uh started with the global launch? Um sometime during the soft launch beta. Wow. How many of you played the game in portrait mode, like this, instead of this? Wow. Okay. Nice. I'm sure you have some questions. I hope this talk is going to answer some of them. Or raise more questions that I can answer later. All right. So, uh the the topic of the talk is a bit whimsical. Uh so, if you don't get it

now, uh well, let's say if you don't get it after the talk, then I didn't do my job. But by core, I mean um some core design element of the game. And uh the talk is about how we during the soft launch, which went on for a long time, tried a lot of things, a lot of very fundamental changes in in our game, and uh what the results were. But first, a little bit about me. I started at Supercell about 8 years ago. Uh worked on the very first browser-based MMO um and also have been working on various fun um small projects, iterations, and uh experiments, prototypes. For example, this one. This is a world premiere for Forest Ranger. circa 2011 um just a quick

one week uh prototype with one artist. And also the uh after we uh pivoted to mobile games, I worked on the very first game that was killed by Supercell. Um RIP. It's uh appropriately in black and white. They're very morose. They're called Project Tower. After that, I worked on Hay Day, Boom Beach, and early on in Clash Royale. And now for the last 2 and 1/2 years, I've been working on Brawl Stars. So, today I'm going to be uh covering these topics. First of all, how we got into beta. Um Ooh. Yeah, right. Okay. How we got into beta. Uh some

Uh and then getting into the meat of what I just was talking about a moment ago, the uh iterations we did trying to find the perfect perfect control scheme. Uh improve um player engagement and experience on the long term with different iterations of the progression system. And then stuff we overlooked, social gameplay, and uh I wouldn't say overlooked, but uh I certainly have a different perspective now on the kind of feedback we got from the community. So, let's jump right in. Um Back in 2015, uh there was a lot of excitement in at Supercell about Clash Royale. The game wasn't going to be in beta for several months still, but um the promise of a real-time uh multiplayer, at least two-player

uh competitive game was was definitely in the air, and uh we thought the world is ready for a game that quite a lot of us a type of game on mobile that quite a lot of us like to play um and didn't exist yet. So, uh after even after uh Clash Royale went to beta, it this was proven, and we thought, "Okay, the market is ready as well, not just uh the technology." And um but even before Clash Royale went to beta, um we were prototyping with a small team on uh a multiplayer action shooter. And uh between 2016 and 2017, we had three internal milestones. We call them uh or the target of these milestones is always

a company playable. Being a a small relatively small company, we have the advantage of getting uh or potentially getting everybody at the company playing our new game projects, getting a very uh wide variety of different backgrounds, from, you know, finance to legal to marketing to game teams. Um player support uh engaged in our our games uh and giving feedback from their own perspectives. Also, um this has been uh kind of a staple of of uh Supercell game development for years now. We also uh tested with real players um because for us the the major worry was um

fast-paced controls on a mobile device or any kind of touch touch screen device. And we wanted to see if our designs control scheme designs were uh actually usable by regular people. We were confident enough to soft launch the game on June 15th of 2017 in Canada and on iOS only. And uh we'll come back to this decision. Um actually we'll talk about it right now a little bit. So a lot of you know why uh game soft at soft launched uh in Canada uh the for Supercell the reasons are pretty much the same. Uh the it's a smaller market that's quite comparable to to um the US, which is one

of our major has been historically been our our most important market. And uh it also allows us to to compare the performance of the beta um to our previous previous betas and get it some insight of how we're doing, what the target should be. It's much develop easier to develop just for one platform. So uh and iOS has traditionally been the one we develop for at at Supercell. For Brawl Stars in particular um matchmaking was a concern. So we wanted to limit ourselves to one geographic region at first to have people playing in the same time zone so that our uh beta population wasn't spread out all over the world to make sure that they have uh good fast matchmaking experience. And finally, we were

uh concerned about uh the game spreading, especially the APK spreading out and uh too many people, though more people than we would prefer having the kind of uh not finished, not polished uh soft launch launch experience of the game instead of saving it for the global launch. It turned out to be the longest beta in Supercell history. 545 days which was a long time for both the players who started early on and for the game team. Uh the reasons for that is are well, uh again historical. We have been able to uh release uh very popular games before and all the new game projects are going to be

uh all the new games are going to be compared to these these games. So, we already had some um fairly uh ambitious goals uh going into the project. And um in addition, while we're quite happy to kill non uh poorly performing games in soft launch uh we're committed to for to support the game for years um for the players uh once we go global. Okay. Now, I'm going to show a quick video. I'm not sure if the sound's going to work or not, um, how the game looked when we launched in June 2017. And if you play the game today, you might notice that it looks quite similar to, uh, well, the fairly similar to how it how the game, um, I mean, the that state almost two years ago and, uh, and today look fairly

similar, but almost everything has changed under UNDER THE HOOD. SORRY, NOOB. I BRING BALLS. BLING BLING. YOU WANT TO BRAWL? ALL RIGHT. SO, LET'S TALK about some of those changes under the hood. First, um, I'm going to dial back the time a little bit before we went to beta and, uh, talk about various control scheme iterations we had, um, before and during.

We, uh, were are still pretty, uh, proud of how we've been able to innovate touch with touch screen controls, um, in Hay Day, in Clash Royale, in in Clash of Clans even, uh, we felt that we really raised the bar to how you, uh, the for the usability of of touch screen games. And, uh, we didn't want to go with the default option for an action shooter, which is um dual joystick type of game. These games were out already, several of them, but basically none of the team, the early team liked um dual control, dual dual joystick controls. So uh that started a long process of prototyping, experimenting uh with the

main objective of trying to take some of the twitchiness, the fast-paced movement, and uh excessive tapping um or control actions out of the out of the game. Uh this is one of the prototypes where the player actually only moves the character by tapping on the screen, and you can see a faint outline uh where uh that shows the range of the characters' um weapon. And once an opponent gets into range, the character will automatically fire. This uh was a fairly com- fun control scheme, uh but we thought we could do better, and also felt like we should give the the players more control. Also, there were always complaints about the slow-paced feeling of all the kind of deliberate pace of the gameplay.

Here's the following iteration. Um and in terms of the uh core combat, this is pretty much what ended up uh in the soft launch version. Even though you might notice the uh the game looks fairly different, but uh that's a that's a topic for a whole other talk. So in this uh iteration, you can see the little cursor moving around. Uh the player taps where they want to move and swipes and releases to fire. And the place where they release um is the point targeting point of their attack. So, now we moved from auto attacking only to pure skill shots. And we thought this felt um like a very uh good compromise. Well, it didn't feel too hectic and the controls were simple and and reasonably easy to learn.

Until we started user testing. Uh this is from um March 2017. And uh an actual user playing the game for the first time, having the um s- about to start their first game. And uh the immediately we noticed something uh strange or something that was different from the way we played the game. Uh so, in the team we'd been playing the game uh by holding the device on one hand and then tapping with the index finger of the other to do the controls uh execute the controls, movement, and the firing. But, this is not how people actually use their cell phones. So, um instead of pecking on the screen like an old person, uh

our testers were using two thumbs as in as if texting, the regular way they use their phone. And also, our ideas of uh deliberate and uh kind of uh low actions per minute uh gameplay went out the window. It turned out to be incredibly hectic, as you can see on the video. So, the the feedback from shooting wasn't sufficient to to uh let the player know that this is all the actions they can take for the moment. So, uh the testers and our uh co-workers alike reported feeling like feeling very overwhelmed by the combat because they felt like they had to do 250 actions per minute. So, based on this testing, um and the feedback we'd been getting,

we did one more iteration uh before going to beta for for our control scheme. And this was introducing the dreaded virtual joystick. So, uh we didn't change the game uh dramatically unlike between the two previous iterations. Where, by the way, I forgot to mention that the reason we turned the game into portrait mode was for the more that more uh natural way of holding your handheld device. We were designing for uh phones as the primary platform, and we felt, especially with the experience of Clash Royale, that this was the way uh way to go. So, there was one virtual stick at the bottom corner. You could uh customize that for either left- or right-handed. Um And then you would tap where you wanted to shoot.

Now, user testing uh found that uh this was also frustrating for for players. And um the kind of drifting stick uh we had designed ended up drifting all over the screen. However, one thing we found from both the company playables, user testing, and our um just playing with the team internally uh was that when you played together, when you were in the same room, um you could put three absolute beginners in the same room, and they would still have fun even if they struggled with the controls. This guy is really enjoying himself um after playing the game maybe about 10 or 15 minutes, and they uh there's good communication going on, and um there is a lot of laughter, and and

uh overall uh a really nice game uh gameplay experience. We'll come back to this later. Okay. So, here are the two control modes we had to choose between. Um tap, once again, you hold the device however you hold it, and tap by your, you know, chosen finger to move, and then swipe and release to fire. Whereas the joystick movement was um basically holding the device on one hand, and then with the other hand free, you would tap where you want to shoot. The joystick would control your character. We couldn't decide the um user testing was indecisive. Company playables uh I mean, voting after our company playable gave us no clear winner. So, we decided that

let's let the players decide. We uh assigned all new players decided decided to assign all new players in the beta uh with one of these two control models. However, even they couldn't decide. So, uh looking at uh retention, how long players stick with the game, uh there was no clear winner between the two control models. That left us scratching our heads until um our data scientist dug a bit deeper and found one clear difference. Players were switching more to the joystick controls from tap controls than vice versa. And that led us to eventually phase out the tap controls. They remained as an option in the menus, but everybody got joystick controls as the first choice.

Now, there's a twist. And it's this. Eventually, about 9 months later, we decided to turn the game back to landscape. Well, back, it had already always been in portrait mode. And this was more or less uh natural progression from that single joystick. We had also already experimented with double joystick, even though we hadn't tested it, but on a portrait mode screen, uh you couldn't didn't really have room for your two thumbs uh to maneuver uh if the phone was in portrait mode. So, with the popularity and eventual acceptance uh of victory of the single joystick, um it just felt like um more natural method of controlling the game. Going to

one joystick was basically opening the door for the two joysticks. And that required this quite big change. And of course it had a huge impact on core gameplay. We had to uh teach players who'd been playing the game, some of them for 9 months, to uh relearn the shooting controls all over. We had to uh implement a whole new uh auto aim or quick fire feature where tapping the right joystick would shoot at the nearest opponent. And uh I'll come back to the the fallout from this in the uh community feedback section. All right. So, these are the different uh control mode iterations we made uh before and during the beta. Let's move on to progression. So, when the game went to beta,

oh, there's another pun. Um the upgrade system well, uh the upgrade screen looked like this. So, we had decided to go with an upgrade system as uh something for players to strive for, to give them meaningful progression from uh one session to the next or one week to the next um when they've they've been playing more and more. And uh our first uh iteration of this that actually went out to players was very much like an say an RPG game where you had points to allocate to different uh character uh traits or attributes. Now, in a competitive game, this is risky because the any kind of power progression means that

uh you'll sometimes face players with more powerful characters than you. So, we tried to walk a fine line here by keeping the power scaling very small. Compared to for example, Clash Royale has if you look at the very maximum of of scaling you can have with common cards, it's about 200%. So, we we went on the very conservative side. And if you take into account the fact that you rarely face due to matchmaking uh the have this maximum difference in any one match. Anyhow, uh it it felt like a good compromise. And in fact, even though we were quite concerned about this choice, uh we had good results. So, players weren't complaining about the game being played to win. Instead, the progression system, at

least in the beginning, felt meaningful. Um They were this uh upgrade resource used for uh every kind of upgrade in the in the game was in loot boxes and also added uh much needed content to the loot boxes so that uh the box we game gave the player at the end of every session or at least end of of every day if you played actively uh would contain something useful that would make you uh feel like you're getting ahead in the game. However, it wasn't very satisfying to get a single elixir bottle. Anybody remember this screen? Yeah. And uh in fact, even getting massive amounts of elixir

didn't feel that great. There was a long-standing uh misconception around among our players that uh getting 10 elixir cheated you out from getting a legendary character. So, people felt really bad missing out on Spike and Crow, or so they thought, even though the drop rates were very very different. But, many attempts to communicate that um the rumor or the misconception just stuck. We saw that people weren't sticking around um and the assumed that that the big part of the reason was uh that the this progression system wasn't satisfying enough. And we also saw this uh from the community, a lot of complaints about getting duplicate brawlers, which was another thing back then, and um and small elixir drops that felt

meaningless. So, we decided to iterate the progression system. So, we 6 months into the beta uh it looked completely different. And uh if you don't get what's going on there, I don't blame you. We uh addressed one issue that people had with the progression system, which was uh having difficulty to deciding where to allocate the elixir, which characters to upgrade. You would always be uh second-guessing which character is going to be nerfed in the coming update, which character is going to be buffed. Am I really going to enjoy playing with this guy? Because once you allocated your resource, it was gone. Well, it was permanently allocated with no chance of of

respecking. We took some of that uh difficulty of choice out while keeping the same amount of power scaling. So, instead of a unified upgrade resource, we now had clip uh these uh pins, badges, medals, and crests, which were unique items for each character um improving one of their stats. And all of these uh were added to the loot boxes. So, there was a lot more variety now that you could get when you open a box at the end of the session. And uh there was even a chance of non-linear upgrades. So, you might get lucky and uh pick up a crest for attack damage, for example, from just a random box. So, now there was a

there was some dynamic range to the boxes you opened other than just characters, which were quite few and far between once you had the common ones and the rare ones. In addition, we went a step further by uh adding a star power to each character, which actually changed the gameplay. So, as you can see, maybe it's tiny text, but uh Shelly Shelly's star power gives her the ability to slow down enemies when she hits them with the with a super. Um there were some some downsides to it. It made balancing between uh characters who do do have the star power and don't have it um a little bit more complicated. You had to take both

uh into account or just balance um for the for the top level uh fully upgraded players. Um and also the having the star powers in the game made the game a little bit harder to understand. You didn't know exactly what you were facing when you entered the game. You might be slowed down by Shelly's super or not. And also the whole upgrade screen was really hard to explain for first-time players. It did its job though. We were able to see uh noticeable bump in longer-term retention. However, uh here comes the twist. We were going to have to redo the UI anyway when we went to landscape. And as a leap of faith, uh

we decided to try something well, tried and true. So, only 3 months later, uh we released another progression system. Yeah, you can imagine how the players felt. Some of you felt it, too. This second iteration uh during the beta uh had a upgrade uh or progression system that was quite similar to Clash Royale. So, similarly to how cards are stacked until you can upgrade them, um now we had in Clash Royale now we had power points that are character specific and you use coins to enact the upgrade. There were some changes because of that.

They were There was absolutely no choice anymore when it came to choosing which character trait to upgrade. You could only pick which character to actually finalize the upgrade for with coins. So, uh to make kind of make up for this change, we introduced 10 different upgrade levels and increased the power scaling uh in order to make each level feel uh significant. So, uh this was another kind of leap of faith to see uh and we had to see how the community would react, but this was the least of the communities' worries. Uh the game still apparently felt fair, but um even though this new interface was much much nicer uh nicer looking and uh

there was no more confusion about pins, badges, medals, crests, tokens, et cetera. It really uh simplified everything. We couldn't improve metrics and uh a bit later on I'll be talking about how the players felt about this. All right. Now that I've gone over the control scheme iterations, the progression system iterations, Uh let's talk about something we didn't improve for a year, which is playing with friends. So, as I said before, we knew the game is fun when you play it with friends. It's That's the best gameplay experience.

But, actually playing with people you know together was uh quite difficult. We made you jump through quite a hoop. Uh you either had to be face-to-face and uh input this five uh character code to enter your friend's game room. Uh or you might uh get some people, semi-random people from your clan that we called bands back then, which are called clubs now. Yeah, iteration, right? So, you had by far the most fun when you managed to play with your friends. And uh for example, our finance team is still playing together uh ever since the the early beta because they got into the habit of

typing in the room code even now that you don't actually always have to do that. Um they they play together and have had amazing retention. It's you can't say the same for most players. Um Even if Even though you get a competitive edge uh from playing with your friends, you can uh figure out the team composition that you want to play for a certain map, certain game mode. Uh for most people you most people ended up playing by themselves just trusting the random matchmaking. And instead of focusing on some way to get your actual friends uh in the same room with you, we had this uh we only enabled players to to invite band members.

And band members typically they aren't in your close circle of friends and basically anybody who happened to be online might join you. There was no in-game friends list so if there was somebody um that you played with, some random person, there was absolutely no way to con- connect with them afterwards even if you wanted to. And finally like I said earlier on this was a a very faithful decision for us to go with iOS only because if you think about a game that it's best that is best played with friends um if your friend if you have an iOS device and your friend happens to have an Android device, you just can't play with them. And that's a hard stop. Um unless you get them to buy

an iPhone or iPad just for Brawl Stars. And that's a hard sell. So word of mouth and uh the kind of excitement a fun multiplayer game can achieve was definitely dampened uh by going iOS only. And also uh you can't really uh tell your friends to start playing a game if yeah. If they can't play the game. All right. Which kind of gets me to community feedback. So let's start by talking about our the way we handle our community. Ever since the beginning of the beta uh it's been centered on Reddit. Even though we have some social media presence and a YouTube channel um the we actually uh direct people from inside

the game to our Reddit with various in-game news posts. And we've had several thousand active more or less active members on Reddit throughout the beta. And we also communicate uh with our community through through our sub-Reddit. In return we get a lot of suggestions, complaints, and all kinds of cool stuff like fan art. However, having a very passionate community um also means that they're going to speak out when things are not going well from their perspective. So all the stuff I've been telling you about the iterations we made to uh core functionality

uh was very jarring for the player community. So uh we changed the progress system even though we always refunded everything. So uh people got back anything they've they'd spent on uh improving their characters and got to respec. It was still a big jarring change to relearn, but even more than that, um, changing the control model was was really tough for a lot of players. They had to relearn, uh, very basic functionality of the game and to give up a lot of the skills they'd learned. And on top of that, we provided very little content between these big core gameplay changes. And, uh, the thinking was that we already know from the early beta that whenever we release something new, a new brawler, a new game

mode, or something like that, that is exciting for players, uh, it increases our metrics. And, uh, and so now we can focus on something else. And finally, uh, as the, you know, hard as it as it is to say, there's often been this kind of drop in the bucket mentality where our beta player base is seen as, um, sort of expendable. Not entirely. We do listen to them and, uh, we communicate with them and appreciate their feedback. However, uh, compared to, uh, the global player base, this is a very small amount of players. So, we take this the beta stage as a chance to make big changes and experiment. And

after global launch, those kind of big changes would be very, very detrimental for the future of the game. So, uh, it's a sacrifice we've been willing to make. However, with an active, passionate, and angry community on Reddit uh, just two taps away from starting the game. Uh where you can see a new player can see a wall of Okay, this is not a wall of posts, but just a complaint about how the the community sentiment was um as recently as uh a year ago. It's a it's a really big warning sign for somebody who's starting a new game checking out the community what the what the game is has to offer. And

even our content creators were affected by the the state of affairs. So YouTube content creators have been really important for us historically and currently um as uh people who who driving new players who keep the current player base engaged and uh overall sort of uh positive avatars uh to show what what our games are at their most fun. Here's one of them. Carrotstime uh has been with us from the very beginning and he's one of our most popular YouTubers. And uh okay, he looks like he's pretty sad at this point. But actually he's acting because we did something different in May 2018.

So this was uh a stage in in the uh in the beta where even the game team Well, the game team was already uh very frustrated uh by the apparent lack of um movement in metrics. We weren't able to improve the games efficiently to feel confident that we could globally launch. And uh so after these major core gameplay changes, we decided to focus on something else. Okay, we now gotten all that stuff off the plate. Even the huge previous update had no significant impact on metrics. So, instead, we released new characters, new game modes, and crucially improved social gameplay. There was now an in-game friends list,

a play again functionality where you could actually keep playing with the same randoms if you found them to be a good team. And improvements to economy, and a new focus on rewarding the player for climbing in trophies, for getting better. So, we instead of just uh im- rewarding the grind, now we very visibly rewarded the players actually getting better, winning games, and moving ahead. Uh and for the first time we kind of reversed course, we took rewards out of loot loot boxes and offered them as rewards for uh climbing in trophies. And then in June, okay, this is the same

lonely spike in the desert photo I saw uh pic I showed before, but actually it's a teaser for Brawl Stars Android launch. There's a if you if I can point your attention to Spike's shadow, that's where the teaser is. Ever since Brawl Stars went into beta, every YouTube post uh YouTube uh video we posted, pretty much every YouTube video our influencers have posted, every Facebook post um every pretty much every uh official Supercell uh Reddit posts on the Brawl Stars subreddit had a bunch of where's Android? Where's Android? Where's Android? That was the number one question every time. We had a lot of followers who didn't even

themselves have an iOS device, but were so where either Supercell fans or just wanted to play the game and uh were committed enough to come back and ask for the Android release for months and months and months. So, this smashed down the wall between uh iOS players and their Android friends and Android owning friends. And we actually still have uh organic growth si- uh ever since uh this update in some regions where we don't actually actively acquire new users. So, it's um quite amazing what a difference it makes when you allow people to play together. We were for a while before talking about how well this game could do in Asia.

And uh so this was planned as a sort of one-two punch Hail Mary shot to, uh, see if if Brawl Stars still had legs. And and could be viable. And yeah, took us a year. And what was the result after June 2018? Hm. Okay. I think I broke this slide. Here we go. So, uh, here's what happened. We'd been very used to hearing, "Thank you, Supercell." on Reddit, uh, in a very sarcastic tone. As you can see in the in the top comment there, uh, "Thank you for ruining such an

amazing game." After the May and June updates, it basically, uh, took a 180 turn and people were super excited and thank grateful for us for, uh, actually us actually listening to the community and provide providing this update. Uh, and with, uh, the numbers in Asia looking looking good, the game team was energized, content creators had, uh, a lot more to to, uh, make videos about and everything looked good to go go global, take that chance. But I'm not actually They're to talk about that now. Uh, I'm going to leave you there and

talk about some takeaways from the long beta. So, first of all, I think we made a lot of nice innovative uh designs. Both in terms of controls and in terms of the upgrade progression systems. And there are still things, you know, people look back on players and and game team members alike with some nostalgia. There's always um a give and take and a compromise when you uh when you decide decide to iterate. And some of the innovations still feel close to heart and might have actually worked if uh you know, if things turned out differently. However, you don't shouldn't be afraid to fall back on on proven designs

as we did both with the controls and the progression. Everybody in the team agrees that this is the best Brawl Stars we've had. The other thing is that looking at retention to uh try and gauge how much fun the players are having, how engaged they are uh has its risks because there's a lot of things that go to such a broad metric as retention. And uh like we experienced uh with with the iterations we made, we weren't really able to make a big dent or big bump with the metrics at any point. The first meta game iteration was uh upgrade progression iteration was probably the best we had. Um

And now this might be a little raise a few questions, but at least for Supercell games a lot of players know that they're playing a beta. So you have to be cognizant of that. So player behavior will change uh if they're unsure about where the the game is going. If you don't release new characters, that might affect even new cohorts when they look back at hey, what's this game like? A lot of players are very savvy and have been there before with with soft launches. Especially a lot of the more valuable players to you, the most likely to engage or monetize players. Then I talked about this. You should implement the most fun features of your

game, the the things that you know are um are the best game experience and try to enhance them and not just believe that okay, we've proven this or we know this intrinsically and we can do something else first and and try to bump other stuff first. Um it These could be such important um This could uh the most important fun in your game might determine the overall game experience so much that it ends up masking other improvements you might make. Now this is an assumption on my part, but uh but I firmly believe that if we'd uh focused on social gameplay earlier, we might have been uh out of the of the beta much sooner or at

least confident that the game is going to fly much sooner. And then finally, uh often when you read uh posts by your community, it can seem that they're uh in an echo chamber just um shouting out the same thing and uh reacting to minor things that you already know or you feel like uh aren't the priority right now. And this does happen. Um but that doesn't mean that the community is always wrong. Actually, the community was quite right in our case. So, it was it was a compass that constantly pointed towards Android. Uh and it just took us a while to navigate there. That's my takeaways.

But wait, there's more. Tomorrow, Frank is going to give a talk on uh that final push. Well, including the the May update, some uh of the team dynamics of how what how Supercell was able to push Brawl Stars out and uh that final push towards the global launch and beyond. So, please go and check it out. Thanks for listening. All right, any questions? Yeah, could you go into the friends changes a little bit? I mean, you said you you said you had room codes and went to friends list, but it was kind of like you went quick over that. Could you go into that a little more detail? Sure. So, uh we uh added

I kind of quickly um mentioned this, but we added a way to keep playing with friends uh uh keep keep playing with uh strangers um who you'd met in the random matchmaking queue. And then uh add the added them to a suggestion list where you could actually add them as permanent in-game friends on a friends list and then invite them easily to your game rooms in the future. Uh yeah, hi. Hi. I I have already every Supercell game on my own test pad. Well, I'm Oh, but Right. the for this this Brawl Stars game, I'm a dead player already. Uh not not your fault, maybe. It's I'm too busy. All right. Uh so, I was playing a game at earlier last year, and I think there is a period actually you have advertisement which can double the loot

boxes. And I I was there and collect a lot of this uh rewards, but something like I think it's canceled. Uh Do you think it's not unsuccessful or it's like you try to find another way to get income from this game? Um so, you you're talking about the the coin or token doublers that we had in the game. it's I think you tried, but you canceled it. Yeah. Yeah, we thought they were they were two different systems that were overlapping at that stage, and we decided to keep just one, the one that was more easily understood by players. You know, not the time-based. We got rid of the time-based one and kept the one that had a very clear you know, system where this this is how many coins or tokens you'll keep doubling in the future. So, yeah, it was

also not that well received by players. We didn't get so many sales. Okay. On the time-based one, yeah. Okay. Um Hello. Thanks for the talk. On my side, I played a lot of Brawl Stars with friends locally because I think it's really important to communicate to play effectively, but you don't have any in-game way of communicating, no emotes, no voice recording. Is it something that you have tested and cut or is it something that you plan to integrate in the game? Yeah, it's another thing we've been talking about for a long time. I think it would go into that social gameplay section that we were very slow in starting to implement or improve. Yeah, thanks. Uh for me, I had a question about the general, you know, vision and design philosophy towards the elder game because it's something I found lacking in both Clash Royale and Brawl Stars that once you get your cards, if you don't want to rank on the leaderboard,

you really don't have something to do and you definitely see that in your in your version of the progression bar where after about a thousand trophies, there's just, you know, a very little content that you get from it. So, yeah, I just want to know your design thoughts on the elder game. Um I'm I'm not sure if I understood. Can you rephrase? Uh sure. So, after you progress a certain amount in the game, and you've got all the characters, maybe you've got a couple of the star powers, there's really nothing for you to do because you're kind of done. All right, so we're looking at ways to give give you more sort of more goals when when you're climbing in trophies instead of focusing just on the stuff you can collect from loot boxes or buying from the shop. So that ever since the May update with the trophy road, that's been a focus for us. And now we're looking at ways to make it more sustainable where it

doesn't feel like you hit the ceiling. So we're looking at different options for that and definitely the the focus is now going to be on the on the trophies next. And obviously we're going to add more characters and add more well, obviously with every character there's going to be another star power. And so the the content will keep coming and also we're looking at ways to improve the the trophy grind as well. Got it. Cool. Thank you. Thanks. Um so thank you a lot for your talk. It was very very interesting. I'm Marie. I work on The Elder Scrolls Blades and it was really interesting to see all the adventures you had with landscape and portrait. I kind of gave a talk about that on Monday and I was and as you may know Bethesda decided because there was no

clear winner for very similar reasons than the ones you said, we decided to allow players to basically flip their phones when they want. So we removed the orientation lock and I was wondering if you had considered it and why you didn't retain the idea. Right. Um so for Brawl Stars being a competitive multiplayer game, uh it's really important that the players are on a level playing field. Yeah. And even now we have to do some compromises when it comes to very wide aspect ratios where the some of the screen is covered. You don't have vision there. If we had both portrait and landscape at the same time, we would basically need to have a square in the middle and lose most of the screen real estate to be able to make it fair and even for for everyone.

Makes a lot of sense. Yeah. Thank you. Thanks. I yeah, I had a question about your early internal playtesting and how many characters and modes you needed to put in there before internal playtesting started to be a lot more fun, you know? Was it Was probably a lot more than one and so the that's the question. Right. So, we went to beta with 16 characters. I think the the early um Well, you saw actually the the character selection screen uh there earlier on But but even before beta cheat. Yeah, so in these are both internal playtests here. We had six characters in this playtest. And in this one we had Let me just start it. Only five. So, um And this was the game mode in most of

the playtests internally. We only added a second game mode, I think, for the final internal playtest. Got you. And one clarification question, it sounded like you were saying that all these changes to the meta and controls barely moved the metrics and it was really the social the social changes that That is Yeah, well, social we can only assume because uh well, even after that we improved the the social game play, we didn't see a massive bump in the previous beta countries. So, I don't know if we burned out our user base what exactly happened there. But luckily we launched in Asia as well and got a fresh new place especially in Singapore where we had amazing numbers compared to to before. So, yeah, one of the progression iterations did give a nice bump in day 30 retention though.

Okay. Yeah. Thank you. Hi, so if I'm understanding your monetization and progression systems appropriately, it seems like a core part of it is unlocking new characters and and then upgrading those characters. Which seems to be a little bit at odds with at least for new players that the the game play of wanting to master a character which for a a skill-based game seems like it would be more of an element than you know, games with with a lower skill ceiling in the industry. Do you guys see that tension as a a problem to be solved, something to improve or are you you happy with where it's at? Um, so the question is about uh the amount of skill cap we built in or Uh the the idea that some players are going to want to sort of master a character, play Oh right, right. Yeah, meaning yeah, okay. Now I got it, thank you. Uh so uh One more question. Okay. So, I'll try to answer this quickly and give one more. So, uh we see this uh happening and it's uh not the optimal way to play the game.

Actually you'll get a lot less uh rewards from and and progression in the game if you do that. So, we're hoping that the system where you share trophies uh uh well, each character has individual trophies and uh it's easier to progress early on with every new character. Eventually will drive people not to main a certain character or focus on a certain character, but try to play the game in a in a different way from maybe what they used to because the game itself has a wide variety of different events and you should learn to play at least you know several four five archetypes of character to be able to be competitive and play in a team game. Hope that answers your question. Okay. Yeah, hi. Hi. Great talk today and we all you know that

you have a soft launch for a long time for these games and we also know that you kill a lot of games before even it come out in your company. And so my question is when it comes to the decision whether continue to make adjustment or just give up and move on to another game. What factor do you consider? So before going into beta it's the team decides if they feel like the game has a chance. Also the internal testing company playables they give some they give very important feedback from the whole company to the team as a sanity check to see if if they're hitting if if the if the company shares that feeling or the other employees share that feeling, but essentially it's the team who decides to go to beta. So obviously we trust it has to be based on trust. So we trust

every team to be realistic in the expectations and if the if the team feels strongly that the game is ready for for soft launch we got the soft launch and then uh uh after that basically the players decide. Thank you. Okay. Thank you very much for coming.