ARK: Survival Evolved: Lessons from the Trenches of Early Access

All right. Hey everybody. Thanks for coming to our talk. Uh, we make ARK. A lot of us in this room also make ARK. Got a some of our team here. Um, we're going to be talking to you a little bit about our journey through early access. And uh, so let's go ahead and uh, kick it off. Uh, I'm Jesse Rapczak. Um, art director, technical art director, uh, co-creative director with Jeremy on ARK. And I'm Jeremy Stieglitz, the uh, the designer, lead programmer, and co-creative director of ARK. So uh, we'll just start out real quick and uh, get right into it and why we chose to uh, ship our game in early access. You know, ARK is a very large game in terms of its scope, uh, gameplay mechanics, and uh, breadth of content. And it's uh, we're a small team, only about uh, 25 people or so on the core team. And it would be very difficult, maybe impossible, for us to

really uh, develop such a game without large-scale testing iteration. And one of the concepts we've really tried to take farther in ARK than in other games uh, we've created in the past is the concept of emergent development. Uh, and kind of maybe a similar concept to emergent gameplay, where emergent gameplay would be gameplay mechanics you don't necessarily uh, plan for or predict uh, as a designer. Emergent development is kind of the concept that we we don't know every direction the ARK development process is going to take us. And those things are kind of determined on the fly based on community feedback and metrics uh, throughout the early access process. So uh, one example of this is really like the uh, creative gameplay ideas. Uh, we really iterate very heavily with the community on this. If you've ever paid attention to our game, uh, we've got a very two-way dialogue. Uh, Uh got a great body of uh, testers and uh, people that give us early feedback when

we're working on stuff and we really implement those community suggestions to make the game better. And and that's really a big part we think of being successful in early access is establishing that relationship with your game's community. And so we'll be talking about that a little bit during this presentation as well to say how we do it and maybe provide some ideas on how you might do the same thing in your title if you're going to go the same route. The concept of emergent development essentially uh in our case means that we have a set of guiding first principles that really define whether something is ARK but we don't we don't have a specific rigid master design document or master plan that defines all of the gameplay features or content that we intend to have from start to finish. And we essentially leave leave that to the iterative process of early access to guide the game

development as long as it fits within this first principles framework and then of course a slightly more refined scorecard metric that we're going to go over later. But I wanted to give you a sense of really what those first principles are in ARK's case. What we're going to do is walk through the original pitch document for ARK and this is effectively all that we have as a design document for the game master design document. And essentially we don't do anything in ARK that would contradict this principles design document but this document is not necessarily a blueprint for everything that goes into the game. So don't contradict but otherwise anything that doesn't contradict is fair game. That's pretty much what I just described. So this is the original design document that we generated prior to commencing development on Arc. And I'm not going to actually go into each detail slide of

it. I'm going to kind of step through it real fast, but you'll get a sense of how high-level it is. And this is as close to a design document as as we have got on the title. What's interesting is while it is very basic and high-level, it's all actually still accurate to what the game's current state is. We've never explicitly contradicted it. And of course, if you go through and you know, scrub through the video recording of this, you'll be able to look at the document's details in in more detail. It's probably worth noting too that the design spec we put together here informed a lot of things. It didn't just keep the team on the same page. It also let us tell our partners, hardware partners, platform partners what the game was about in a very succinct, easy to digest way. And it just made it less complex to understand the scope of the game we were trying to make without getting into all the little nitty-gritty details. You know, and effectively this mapped out a set of axioms that defined what is

Arc as a game. And it it it gave us kind of some guard rails, you know, when we had some ideas that maybe seem like good ideas, but they fit strongly outside of this framework, then they would kind of be rejected out of hand. But it gave us enough freedom to iterate and kind of creatively brainstorm within this framework. And you know, as you can see, it's really more set of goals than a specific list of content. And that was the whole thing. That's all of Arc's design document. But out of that, we've added so many features with the help of the community. These are all actually features that explicitly came out of community suggestions and community iteration. And we're going to go over shortly with how we kind of distill you know, when we get a bunch of suggestions on our forum or on Steam, how do we actually distill that into what we decide to

implement in the game since we obviously can't put in everything that's suggested. But as you can see, many things that run the gamut from gameplay to UI to meta-game mechanics. So I'll talk a little bit about the community. As Jeremy mentioned, we get lots of good feedback, gameplay ideas directly from our community. And we treat the community feedback in the same way we treated our own feedback internally when we are first making the game before we launched. It's kind of a bottoms-up approach, very playing the game. We already know what the high-level goals are. So when we're playing the game and thinking about features and listening to the way the community is playing the game, we're thinking about all those things Jeremy just mentioned with does it fit into our original plan? Is it something that we feel makes the game better. And we've got a bunch of metrics around that as well. But the first thing is really to have that community in the first place. And if you're going to do an early access title, it's really one of the most important things. When you go to launch your game, you better be ready with community management and take it

very seriously. Our community managers are very very good at what they do. They have a pulse on the community and they reinforce the relationship we have with the community on a daily basis. I don't want to make them blush. Some of them are in the room here. Um so one of the first things we feel is really to get the community excited and to keep them excited because hype can be really good. We do this through combination of patch notes. We're always writing up what's coming so people know what to look forward to. Might not always be exactly when it's coming, but we kind of leave that stuff up there so that people will kind of know down the road like a mini road map a little bit of what they can look forward to in in coming months of of development. Teasers we do all the time. Uh when we first launched ARK, we started putting out uh dossiers for all the dinosaurs that would be in the game, even though some of them weren't coming for months or even years. Uh players still got to see the dinosaurs and the creatures that would be in the game, hear a little bit about their abilities and the things that they might do, uh and also give them an idea of kind of

the scope that was like ahead of development. You know, it's one thing to read about words on a page, the Steam page, about what we're going to put in the game, but when you start seeing visuals and start seeing stuff early that's it's real, you know, you start to feel a little bit more uh attached to it and excited about it. Um announcements, too, to go right along with those teasers. Whenever we have something new that's going into the game, we really spend a lot of time putting together those announcements, uh working to get out media alerts and press releases about the content that we're going into the game. And it's not just for content, it's for things that we do around the community that we'll see here in a in a couple more slides. Uh so touching on that, spreading the word is really important. How do you get the word out about, you know, all this hype that you're generating? Uh I mentioned media alerts and press releases. It's really good to, if you don't have a PR company or a media company, um to uh partner with one. They can help you get into uh voicing your stuff that way cuz we all spend a lot of time making games. We're not PR experts. We don't

have the relationships with the media and all that stuff to do that. So it's really beneficial to not spend a lot of time on that, but get it's partner with somebody that can help you maximize your time around those alerts. Uh we always do trailers. Like I mentioned, visuals are very good. Trailer screenshots, presentations for everything we do. We really want players to get super excited about the stuff that we're putting out, whether it's coming right now, which is usually the way we do things. It's like, "Here's the trailer. It's available today." which is very useful uh for getting the word out because everybody's so excited about it that they want to play it right now. Uh so we try to pair those announcements and the media uh with gameplay stuff available right now as well and it's it's probably a good practice. Um so we post stuff on our website survivetheark.com. We mirror that on Steam and Reddit. We try to foster communities where players congregate. You know, we love our website. It evolved in similar way to development where it kind of emerged out of a need for a place for players of all

platforms to go. But a lot of people like to chat on Reddit. They like to chat on Steam. So we try to like mirror our messaging there and spend most of our time moderating our own site. But we go out on social media and the various console platforms, you know, PlayStation and Xbox have their own ways to talk to the communities. Again, everything I'm talking about here is community management. So going back to the first thing I said, it's very important that you have somebody or multiple somebodies that are going to focus on this so you can spend time developing your game and all the great work you do gets out to the community and they get very excited about it. So we want to keep the conversation flowing with our community. We've got a variety of ways to do this that we came up with with our community team. One is one we call community crunch and this is something that Jat runs all the time and Jen and it's just the latest cool stuff from the community. I mean Ark is a game with emergent gameplay about creating stuff. You see all sorts of things from music videos to funny cartoons to

screenshots and media that people are taking and we encourage that because people have a lot of fun with their creativity. We want to highlight it when it's awesome and amazing. Sometimes even if it's questionable, but it's still awesome. We put that out there. We put it in our spotlights. Anything anybody sends us, we've got fan-made dossiers, screenshots. It really makes the community feel like we're listening to them, which we are and we're valuing the work that they're doing inside of our game that that's so important to them and it really creates that kind of like uh uh communication pipeline we talked about. And we also do some content sneak peeks with uh within the Community Crunch. Occasionally, we'll put out a little teaser or something that's outside of our normal, you know, template for for doing that sort of thing, like a a little like mysterious graphic of something that's uh hinted that might be, you know, I like We did that when we were launching new creatures that hadn't been in the scope of the game before. We put like mysterious dossiers in the Community Crunch and stuff like that. So, hiding little Easter eggs in our community

stuff. Um and then we do smaller announcements with Community Crunch, like things about sales, merchandising, gameplay tweaks. Again, this is all community stuff. Very important to have somebody that's like on top of this. ARK Digest is really great, too. Uh usually, this is uh Jeremy um answering direct questions from uh uh community members, players. They submit them all, we cull them, we answer their feed or answer their questions, uh listen to their feedback. And again, it's that two-way communication with the community. So, we're not just telling them stuff. We're listening to them and responding to it. Another thing we do is rewarding our community for all of this great two-way um relationship we have. Uh we really try to give back to them. Uh we call them out in all of their the weekly spotlights with Community Crunch. We run contests and prizes. Uh we reward people for building beautiful things in the game. We give them things like Steam gift cards, you know, a couple bucks here, call outs there, free keys here. We run mod contests. We partner with um

uh partners like Nvidia uh with new platforms like Ansel. We integrated that. We run some contests for that. And we just now launched a sponsored mods program, which is uh kind of a middle ground between being a professional game developer and being a hobbyist. Uh there's not really a way for people to reliably sell their mods right now uh out there, but ARK is uh a game that has a lot of mods, a lot of high-quality mods. Uh this video here is our first uh sponsored mods that we're starting this month and we're literally uh funding these people to keep making their great mods because we don't want them to stop or give up because, you know, the pressures of life or their job or whatever are too much and they have to stop working on it. Uh we really want to reinforce the work that they're doing and the best mods uh will rise to the top and month over month will continue to find those that the the players they're playing and and that they're really loving. Uh a couple serious stuff, too, when it comes to rewarding people, you know, we've in the past done exploit bounties,

uh making sure that uh anything that's kind of uh a hacker or some sort of um thing that's detrimental to the game, uh we can try to reward people for finding those things instead of exploiting them for their own personal gain. Uh just goes along with this whole like two-way two-way street we're talking about. So, the result of all this is that you have a community that's passionate, which we do. Uh they're entitled. They They know that they're part of development and they want to continue to contribute and they've got kind of an expectation from you and you've got an expectation from them. Um they've got a voice and they've got all these ideas, these great ideas sometimes, sometimes not the greatest ideas to contribute back to development. So, that's the big question. How do you know? And you shouldn't always listen to your community and what they are saying. And you have to decide all these emergent features we talked about earlier kind of went down a a path and throughout development we talk we thought about how

do we call these and how do we take these and find the best ones and actually implement them. And that's what we're going to talk a little bit about now. Get through this one. I mean, ultimately uh the community really does have a bigger, more creative brain than any of us could possibly have by ourselves, um, but it also can pull the game in a thousand different directions. And, you know, the loudest voice in the room isn't necessarily representative of the larger player base. And, uh, to the extent that, uh, it's the loudest voice in the room, it might also be an impossible idea. And, for various reasons, it's important to have some kind of metric or scorecard, we found, to kind of weight ideas against gameplay ideas, whether they're coming from us or even coming from the community. We need a way to gauge their viability and desirability, uh, that was beyond just how vocal

people, or even ourselves, were about doing a specific idea. Um, this was particularly important because the game was live. So, any misstep, you know, ultimately would cost us in terms of, uh, players, uh, experiencing the better or worse that we might put into the game. So, ultimately, we came around to implement a feature scorecard, uh, into our internal design process, uh, that would allow us to grade, uh, potential ideas against a variety of criteria. And, I'm going to go over how the scorecard works. Okay. So, uh, it's a series of points. Um, and ultimately, there's a threshold, uh, over which a specific idea must reach for it to be like automatically included in the game, that we should totally do it. Um, and the first kind of question is, do we want to do it? Is it Doesn't necessarily mean that we came up with the idea. That's irrelevant. Um,

but is it something that we, internally at Wildcard, are really passionate about? Um, easy for easy enough for us to determine. Uh, just ask the team. Uh, then, do we think the community wants it? Um, and and again, it's it's not necessarily simple as saying we should do everything the community wants and nothing the community doesn't want. As we're going to see, some things the community wants turn out not to be so great, and something the community doesn't want, at least initially before it comes out, say they don't want, actually turn out to be quite useful to the long-term health of the game. But, we try and gauge whether the community is going to want something. That's usually by putting it in patch notes, by telegraphing it on social media, and and and having our very talented community team gauge the response through a variety of metrics. However, this does tend to boil down to the loudest voice in the room. So, we That's why it's not the only criteria, or it's not necessarily weighted any higher than our own voice in the matter. The another question is is it good for new players? If you just have downloaded

ARK today, you're a new a noob on the island, starting out, is this going to help you get into the game? Is it going to be neutral? Some things we add that are actually very useful can can actually be negative towards a new player in the game. You'll see one of those later that's actually a good feature overall for the game, but it's not necessarily helpful for new players. But, generally that's that's a criteria we want to bias towards is things that are good or at the very least neutral for the new player experience. Is the idea beneficial to development? What that means is is shipping the game specifically. You know, I'm not talking about whether it's good for marketing purposes or anything else. I'm talking about does this feature is this feature in line with the ideas of what ARK needs to have to ship or does this idea kind of sit outside that rubric? Maybe make the game take longer to ship, make it more complicated to ship, add a lot more work to what would be the total content set of the game.

Does it lack controversy? Obviously, we're not afraid of controversy at Wildcard, as you've probably seen. So, it's not weighted as highly as is other uh But, nevertheless, uh it's it's still something we try to factor in. Um is this going to be controversial or is it not? Um And and and we we gauge this, by the way, again by reading the uh community posts and taking polls. Uh it's worth noting that this is a little different than what the community wants. What we look for in terms of gauging what the community wants is whether there is any sizable group of players out there saying "This is really cool. This is a good idea." Conversely, is whether there's also a sizable group saying no. Um is it easy to implement? We we, you know, like any human beings are sometimes afraid of challenges, so we don't like to necessarily take on things that we think are going to be very, very difficult for us. Um but it's not the, you know, most overriding uh determining

factor, and usually the producers will give us a pretty good estimate of what it's going to take to make it happen. Final criteria is, can we do a good job with it? Um we, you know, don't necessarily love to jump into tasks that we think inherently are going to break our team technically or creatively, that we think we're going to do a a poor job at, and sometimes uh that's inevitable, and we just have to give it a shot anyway due to, you know, the greater good. Uh but other times that can be uh a factor in deciding whether or not to go into doing a feature if we think it's just beyond our capability to do what we would be uh proud of with it. But then what we do is we take the results of all of these uh criteria, and sometimes it can be a a neutral uh {slash} zero points, um or a check mark would be the positive points, or a X would be negative points,

and we uh add it up. And we consider our threshold to to two. If it's two or above, we should do it. Less than two, we probably shouldn't do it. And um it's really just that simple. And we kind of run this through every new feature now that comes through the pipeline. And uh There we go. Now we're going to talk a little bit before we get into how this scorecard actually applies to some of the features in the game, let's talk about a subject which many PC developers are probably familiar with, which is Steam reviews and how we think that actually relates to looking at the feedback we get from some of these features. We don't consider it to be the overall metric that is most meaningful, and here's why. Yeah, it's worth noting that uh Steam actually does encourage users to provide any type of feedback they feel is relevant to uh a game's uh development, not just based on a game's merits. So, you'll find that uh this kind of biases the recent reviews portion of Steam to

uh what's been going on with the game in the last 30 days or so in a lot of cases. And uh we've noticed we've gotten some review bombs, and they can be very informative as kind of a response to your actions. And this has been not just for things in the game, but for things that have nothing to do with the game just the development itself. And because of that, it's kind of a little bit to be taken uh in consideration with the other factors you use to judge uh the health of the game. And one thing is uh with the with the reviews themselves, it's it's really just generally another metric for temperature. Um you know, Jeremy, you've got probably some thoughts on that, but Steam reviews, in our opinion, are very, very helpful for gauging the feeling of your community. And it is important to take that into consideration, but they're not necessarily as useful for gauging gauging the overall health of your game. In my opinion, the most important metric for that, in terms of Steam at least, is

concurrent user count. Players. Who's actually playing our game and how many of them are there? Uh we've hit some of our highest CCUs ever, six digits, and we've had some of our lowest Steam reviews ever. And take from that what you will, but uh our our goal is that people are playing the game and they are staying in the game. And that ultimately, hopefully, means a fun game, but we are not as obsessed over our Steam reviews as we are about uh player engagement in the title. Uh now we're going to look at some historical examples uh about how the scorecard was used or not used uh relative to features in the game and ultimately, what happened, you know, what did the data post launch of a feature actually show if if we can gauge it in terms of the impact and does it match up with what the scorecard uh would predict. Uh first, we're going to talk about Survival of the Fittest, which uh I more or less said, "This will be a great idea. We have a quote associated with each feature." Um in fact, the scorecard

itself was designed in response to the failure that was Survival of the Fittest and a subsequent feature. We eventually came to the viewpoint that we needed a system to actually make such decisions. You know, we couldn't go solely based on our gut. So, this one's on me. Um in terms of the scorecard, I'm going to briefly walk you through how kind of So to Stacks up. Um we wanted to do it. Uh it's fun. We we we we we really like the game mode. Uh we thought it was a worthy shot at the you know, trying to do some e-sports type gameplay. Frankly, what we were getting from the community was they wanted to. Uh we we had a lot of fun running tournaments. We got a lot of great feedback that this was something players were really enjoying. It was it was So to by the way, Survival of the Fittest originally began as a mod for ARK that we created as a test and the feedback there was I mean, it was super popular mod, you know, hundreds and hundreds of thousands of downloads, lots of uh people viewing these uh play sessions. We still to this day have

people who are really, really hard core into it. We thought, you know, very much based on what we could tell the community uh was enthused by it. I think they were. Um it's neutral for new players. I mean, obviously, it's not what a new player is really caring about when they uh when they buy ARK: Survival Evolved. And that probably should have been the first warning sign. Uh it's not beneficial to development in the sense of shipping ARK: Survival Evolved. Also another warning sign. It It didn't last controversy because there even though the community wanted it, it was still controversial. Remember what I was saying about In terms of engagement, if the community wants something, we're just looking for, you know, a core of really positive feedback. But so it certainly was converse to the extent that should we be, you know, utilizing development time and development resources for making a mode that is not really considered to be a core feature of the game. Should uh that really be our priority? And that was the controversy justified uh from that standpoint. Uh it certainly wasn't easy to implement. We ultimately uh did spin it off as an entirely separate free product related

to the uh ARK: Survival Evolved game. But um you know, that's a lot of work to launch a a completely separate application and uh manage that and QA that. Um very uh significant time investment there. And uh as far as doing a great job, neutral at best. I mean, while the game mode certainly was very fun even in its final form, Wildcard is not an esports company and ultimately could not uh build the systems fast enough to support esports level gameplay uh with the full tournament and matchmaking systems that people expect of that genre at this point in time. So, uh you know, best case scenario, we couldn't do a great job uh relative to player expectations. So, you know, ultimately, where this gets us is uh we we, you know, perceived that there was a demand for it. Uh there was obviously a great majority of players who at at best didn't care uh and it was costly to develop.

Uh the CCUs were not impressive to say the least. That's the CCUs chart right there on the bottom, by the way. Ultimately, a failure. Um and at the time, we did not have the scorecard. So, had we, you know, quantitatively looked at this, we hopefully would have made a different decision about whether to go forward with it. Ah yes, my favorite quote. The fact that Studio Wildcard still in business is a shame. Scorched Earth, as you may know, is a paid DLC launched in early access. Hmm, don't recommend that anybody really does this. Uh but, um for us, it was a success for a lot of reasons and uh we'll go through that right now. So, uh we wanted to do it because it was important to our development. It was part of our pipeline for creating new content, new maps that players could go to, and it allowed us to really uh develop portions of the game that we needed to do uh for launch. Um we uh had kind of a a huge controversial

thing happen when we launched it, where of course, people were upset that any early access game could do such a thing. Um so, we did get a lot of that sort of pushback. Uh and there was a big review bomb as a result of this. You can see that's our reviews chart on the top there. That's when we launched Scorched Earth or Scorched Earth. Those are the reviews for ARK. And uh it was pretty bad for a couple weeks of of negative reviews. Uh but, it was still number one on Steam for over a week, the expansion pack was. Uh the game the base game was number two on Steam at the same time. The expansion pack was highly rated uh higher than ARK had ever been uh 82% uh and there was also a large increase in uh the daily user counts. I mean, people were loving the content, they were playing it, and the the user accounts were just going up. Uh Uh so, for all of these reasons, Scorched Earth for us was a success. Players loved it. It was a financial success. It helped us uh get closer to the game's development. And even kind of

knowing what we do now about launching, we'd probably still do it because it was such an important part of our road uh to launch. The uh next thing we would apply to the uh scorecard uh is the dino rebalance. And this ultimately uh had us going through and um I like that quote. Uh this ultimately had us going through and fixing some serious magic numbers associated with how powerful I mean, what is that dinosaur even doing? It's killing like humongous giganotosauruses all over the place. So, we we needed to rescale how powerful players' dinosaurs were. And this is always a tough challenge in a live game because players like candy. Um and when you take the candy away, they can be pretty, you know, vocal about it. Um but sometimes that's necessary because too much candy can make someone sick to take that metaphor all the way through. Um so, we you know, we wanted to do it. We we we knew the numbers and how they were scaled in terms of these

tame dinosaurs becoming god-killing machines uh was unsustainable. The general consensus from the community prior to releasing this uh was they did not want their dinosaurs to become weaker. Um it is good for new players because ultimately uh people's super tame dinos were like unkillable for new players. So, it gave new players more of a slightly more level power ramp relative to people who had already gotten tame dinosaurs. Uh it was beneficial to development in the sense that uh we were trying we're trying to shape the game a little closer to its final balance and in so doing this. It certainly was not uh without controversy uh prior and post its launch. Uh there were many people who said that, uh, they would never play ARK again and then they played ARK again. Um, but, uh, it was easy to implement. Ultimately, we're just talking about changing some numbers behind the scenes and because it was easy to implement, we could do a good job at it. It was not the most, uh, technically difficult thing. Um, but, uh, there we go. I mean,

certainly, you know, there was a bit of a review hit for that and like many things we do, but the ultimate data, as you can see in that graph there, was very clear when it came out. The player counts went up and they never went back down. Um, so, you know, sometimes this is one of those cases where the community, uh, prior to its release and in the immediate post of its release was telling us, "No, no, no." But, ultimately, uh, was helpful for the health, the long-term health of the meta game and, uh, I would say that many players, probably most players by this point, have, uh, seen that I think rescaling those numbers had has given the game a longer tail and and have probably appreciated that in the long term. Um, and I I guess the general rule kind of I would draw from that is sometimes you have to do what's necessary for the long-term health of the game in spite of, uh, maybe some short-term negative consequences. Swamp fever is the worst thing I have ever experienced.

That's a good one, too. Uh, so, this one was the other thing Jeremy mentioned that kind of made us realize we need this sort of scoring system. We really wanted it. Um, it was kind of part of our original plans to add communicable diseases to the game and to really kind of make that survival aspect a little bit more hardcore, make you have to deal with a little bit more stuff. The community, uh, some people liked it, some people didn't like it. Um, they probably didn't know what they were in for, uh, before they, uh, had that feeling, though. Um, what ended up happening was, uh, we implemented this feature and it turned out to be very bad for new players. It was the the requirements to construct an antidote were very high in the game. It was very easy to catch swamp fever. And so the new players were just getting like wrecked by this feature. And you know, the players that are already good at the game were like just super annoyed by it cuz they're like, well great, now I have to make more antidote. And so, you know, it was kind of like something that was

we felt was beneficial to development cuz it was getting us closer to our goals. But it ended up being super controversial. Even though it was easy to implement, there is like we really didn't do the best job doing it in the end retro retrospectively. And the interesting thing about swamp fever was that there wasn't really a huge outcry of like this is terrible even though that quote literally says that. But the the community did hate it, but they weren't responding in some of the other ways like that we weren't getting like hammered with negative reviews. You know, there were threads popping up and like oh this is bad. This is I can't believe Wildcard's doing this whatever, but it wasn't like crazy. I mean, we always see that. Anytime you do something, it's the worst thing ever. But we did notice there was a steady downward CCU trend. And for us like that's very strange. And so the only thing that we could conclude was that swamp fever was pushing players from the game. New players were coming into the game, getting swamp fever, and just quitting cuz they would just die over

and over. The the players that were more advanced were just like, well, I'm going to sit this out until this whole thing gets figured out cuz I don't want to like ruin my character and all of my equipment. Wait out the apocalypse. Exactly. So we really miscalculated the effect this would have on everybody, especially new players, but it just made the game very difficult to play. And this is definitely a failure. And this because it was kind of a sneaker failure, we're like, how do we how do we tell in the future if if something like this comes up? And it really helped us kind of design this scorecard around the things that are most important in decision-making and it in our scorecard it's something we shouldn't have done if we had used the scorecard on this feature. This one is a feature that allows people in ARK to transfer their progress between servers. Science-based. Yes. Like most things in ARK and it definitely is uh not conversion

was a a bold move on our part if I don't say so myself. So when we look at the scorecard though it's actually quite directly and dramatically indicative of we should do it and we wanted to do it because we thought this would make the meta game much more interesting. We we like dynamic meta games. We noticed a lot of these prior to this a lot of the servers were becoming ossified, static, the alpha tribes, the big dogs were just sitting on a server and nobody could touch them because anytime someone would join that server they'd squash them like a bug and what cross-ark lets you do is if you're a big tribe on other server you can warp in all of your powerful stuff and you can really have alpha tribes from multiple servers going at each going at it in a dynamic way and so you don't get this kind of really static completely frozen balance of power on a single server. You can also have tribes multiple tribes from different servers team up to take down a super tribe and so we wanted to do it we we thought it

would have some some great long-term benefits the game. The community was apathetic at best. They you know there were a few people maybe who thought it was a good idea but they didn't really no one was really like yeah tile the servers together. It's not really good for new players let's be honest about this because if you're a new player just trying to start out on the ARK you don't really need and you're just getting your little thatch hut set up you don't really need some like gods from above just swarming down and deciding that they don't like you on their beach and wiping you out. So, it's not really a new player thing. It's more of an advanced player mechanic. So, not really helpful for that, but this is one of those cases maybe where the long-term health of the game isn't explicitly uh oriented towards just the new player. It is beneficial to development uh in multiple from multiple reasons. One, we're we're interested in having a long-term uh tail for the game, but two, we also needed this technology to better support DLC in the future. We wanted to allow players to progress their characters from one server to another to enter DLC

servers without losing or having to losing their progress or having to start uh over from scratch. And so, this technology was necessary to support that. Therefore, in line with our long-term technical goals. Certainly didn't lack controversy though when this rolled out. There are many threads on all of our channels uh popping up. There was a lot of people on both sides of the fence. So, we we we certainly think uh the community was was and is split over it. Uh it was fairly easy for us to implement given we actually had already built the technology for this uh for DLC. We just had to kind of wind it and change a few back-end things to allow it to operate between all servers and not just from a non-DLC server to a DLC server. We figured we could do a pretty good job at it uh given that um we had already kind of experimented with the technology for DLC transfer. Um an interesting thing happened when we released this uh feature. Um the meta game did improve. The people who thought Ark had just become a frozen static, you know, wasteland of alpha

tribes sitting on their unassailable fortresses found that they actually had to compete to survive again, no matter how powerful they thought they were. And over time, to this day, our CCUs have continued to trend upwards as a result uh of this feature probably primarily. So, uh by the metric we consider most important, people playing our game, uh we considered it a success and the scorecard bears that out. All right, this is a little embarrassing. I'm sorry I have to talk about this. Um, I would like to apologize for our sheep gate scenario. Uh, in case you missed it, this was a little controversy that arose while we were all on vacation and during the holiday time frame, of course. Um, but basically, Ark was nominated for a Steam award and it was best use of a farm animal. We thought, "Hey, we don't really have any farm animals, so you know, it might be cool to add one if we won. That would be cool, right? Like maybe, you know, people would like that and you know,

they might want to like vote for the game." Yeah. Guess what? People felt pretty bribed, apparently, by that. Um, and news articles left and right. Wildcard's coming to deliver bribing players for a Steam award. Mind you, the Steam awards are very silly. Like if you don't know about the Steam awards, they're very silly. And we had a really quick onset of negative press for us, you know, wanting to give back and put something in the game. And we got kind of a mild review bomb from this, but you know, we just thought it wasn't really worth it. Like if you go through our scorecard, like we wanted to do this cuz we thought it was a cool idea. The community didn't want it because like they didn't even know about it. And when once they heard about it, you know, they definitely were upset about it. We kind of like didn't matter to new players at all, you know. It didn't really help us get closer to launch. Pretty much like there was no reason to do this except we thought we could do a good job at it and we, you know, thought it was kind of fun.

Unfortunately, there was like no real direct positive effect from this whole idea of putting a sheep in the game and it was not really doing anything to help us in a time when the game was already doing really well. Like the CCUs were going up crazy cuz of the holidays and you know, the sales and everybody's getting into new games. And this didn't really add anything to the community or the game. So, we kind of quickly realized this and adjusted our messaging and decided, you know, we're Look, regardless of what happens with the Steam horse, we're going to put the sheep in the game anyway cuz we really thought it was fun, and why not have farm animals in ARK? Um so, it was a failure the way we tried to approach it, uh but we quickly fixed it, and I think we turned it into a win cuz now we have cool sheep shearing and growable hair and all sorts of stuff spawned off from this sheep thing. Um but we probably would have handled it slightly differently uh if we were to kind of rewind and rethink things a little bit. You know, applying the scorecard isn't always uh about uh looking at cases that

are, you know, historical. We also, as we indicated now, use it going forward. Uh We did just launch a feature, so I guess this is technically a historical case as well, but we'll go through it anyway. We want to apply the scorecard to one of our most recently launched features, uh which is called tech tiers. And tech tiers is basically a set of sci-fi items uh for the game uh that pretty much dramatically change how the game is played in the long term. Um and uh you know, we really wanted to do tech tier, the sci-fi stuff, because it was always in line with our vision for ARK. I don't remember if it's in the pitch document, but I'm pretty sure we imply there's a real serious sci-fi aspect to the game. We viewed ARK from the beginning as starting with cavemen and ending at Halo. And uh so, very much uh you know, completing our vision of the players' power progression to the end game. The community uh is is for this. I mean, they think it's pretty cool. New players, eh, it's a really end game type of mechanic. Um look, we could

ship without it uh and still have what we consider to be a complete game. So, it's not really beneficial to development. It is controversial, though. It's uh it's a feature which if you're just interested in this primitive kind of survival, maybe actually diminishes the feeling of being in a primitive lost world. Of course, we already had guns made out of raw materials, so it's kind of debatable how realistic our was prior to the introduction of laser pistols, but I think that players did feel that maybe it diminishes the realism of the title. And it's It's not easy to implement. It's a lot of content that is actually fairly orthogonal to the mechanics of the main game. We're talking about laser guns and force fields and cloning chambers and teleporters. These are all like very distinct pieces of functionality that don't really overlap with the rest of

the title. So, from that standpoint, there's some overhead, not easy to implement. However, the team's really engaged with it artistically and creatively, so you know, we're not considering an afterthought. We We do think we can do a good job at it. So, it ultimately hit the uh you know, criteria of we should do it. And sorry, here's all the things I just said. Uh but here's the interesting thing. So, it's a should do it on the scorecard, and about a week after we launched this feature, we hit our highest CCUs ever, six figures, over 100k players playing the title on Steam simultaneously. So, you know, is there a direct cause and effect relationship there? I think likely. The addition of this feature provides content for the hardest of the hardcore players, the end game players who otherwise might not have anything to do in the game anymore. And you know, targeting them with some long-term content and mechanics it was part of our goal with this and probably brought a lot of those players back into the

title. So, we we consider that a success. So far, so good. So, underwater bases is a feature we haven't launched yet, but everybody knows about. Um and it's something people are really excited about. As you can see, this person's been waiting for years to for underwater bases. Note that Ark has not shipped years ago, but uh just small minor points. Um so, we say we don't really want it because um it kind of adds a lot of work to our plate and it kind of changes a lot of dynamics about the game in a way we'll have to deal with uh for development. We definitely know the community wants it. There's no doubt about that. There's a lot of underwater ocean space in Ark and players want to make use of that space for sure. Uh there's kind of like meh neutral on new players because a lot of the underwater content and especially the tech tier is geared towards the end game, but new players are in tribes and you know, they can team up with players that have underwater bases and there's nothing preventing them from engaging in

that content. Um and you know, beneficial to development. Honestly, we could ship the game without it. Um so, it doesn't really help us get closer to launching that regard. Uh we really feel like it straight-up lacks controversy. Like nobody's like, "They better not put underwater bases in this game or I am quitting Ark. Ark is dead to me." That is not what anybody's saying. Um but it is really hard to implement. I mean, it's totally new system, you know, we have to deal with like physics changing and there's oxygen and then there's not oxygen and then, you know, there's all sorts of collision problems we're going to probably have to deal with and again, it's another place where people can build bases that might get modified if we update maps and things like that later, introducing more bug cases. Um and you know, it's like honestly, something we're like a little hesitant because of all those complexities. How good of a job can we do with this? We're going to do our best uh at it as we always do. Um but just going into it, we're a little like, "Man, this is going to be a lot of work." And you know, like it's hard. Like to be honest, like it's

going to be a hard feature and it's not because we don't think it'll be good for people that we're hesitant to do it. We really think it'll be awesome, but you know, it's just we're human, and stuff's hard, and it's getting close to launch. So, but we already publicly committed to this. No way around that. Community really wants it, and there's kind of this whole risk-reward thing. It's not like survival of the fittest where, you know, it's high risk, and if it fails, it's like really not the best thing. But, this is, you know what? If it doesn't work out perfectly, if it's got its issues, it's a little bit of a less impact on the game in a negative way, and it's another feature that we can add, and, you know, it's not something that we're overly scared about being really bad in any way. So, we think that the we're going to take a leap of faith even though the scorecard says we should not do underwater bases, we're kind of going to ignore that. And that's the good point about this whole system is that sometimes you just

have to go with your gut on something, and even though this feature is something that we probably shouldn't do based on our own scoring system, we're just going to do it anyway. And it's the the risk is not great enough that it's going to kill us if it ends up being a failure. We'll see where we end up in a month. Yeah, yeah. I may be saying something different after like 100-hour weeks of no sleep for the next 4 weeks. So, anyway, yeah. Just to recap. So, we have an interesting uh kind of overall process for making decisions in Arc about what we do. We, first off, keep the community really healthy and engaged through weekly updates, daily social engagement, and a two-way communication process. Uh we essentially don't stick with a rigid static top-down road map. We go with more of a bottom-up approach um to uh designing features in terms of uh soliciting community ideas. And uh but

we don't just say, "Hey, if the community wants it, it's something we should do." We run it through a process now uh that uh includes a grading system. But we also don't religiously adhere to that, either. We we still leave for an element of serendipity, of inspiration, and of faith to uh get ultimately to the bottom line of what should go in the game and what shouldn't. And um you know, if that uh if there's an element of risk in that, then there's also an element of reward. Um you know, overall Ark certainly is a successful early access game. Uh the the numbers uh kind of speak for themselves. Um but uh it's definitely a something you have to keep in mind by going this route. You are best off, in my opinion, not committing yourself to a static, you know, mindset of here's the game you're going to make, and darn if anybody's

going to tell you otherwise. You realistically need to be prepared to swerve left and swerve right um based on how the community is perceiving uh the direction of your game. Essentially, keep an open mind and uh be very reactive. It's really important to uh prioritize content in early access. Uh you'll see some games come out with, you know, decent amount of content, and then really right away focus on, you know, bug fixes and performance stuff. And that's all great. It's an important part of development. Um but we really feel like uh the optimization and non-critical bug stuff is less important than adding the content to the game that keeps people coming back. It keeps new players uh coming to the game because there's new stuff. It keeps old players still playing the game because there's always new stuff to come back to. And the way we've chosen to do early access is throughout the rest of development we kind of front-loaded all the content features and planning to do the final

optimization and non-critical bug fixes towards the end of development. A lot of people take issue with that in the player community. Everybody has a bug that you know they is the most important bug, but in the broader player base that may not be the case and we really try to prioritize which bugs actually get fixed and which optimizations we do because all that work takes away from adding the new content to the game that that we want to add. And it's important to note that nobody buys what they can't see. So it's really important to keep the game visible. Anytime you can take advantage of a sale, do it. Combine it with a content update. It's kind of our the way we we roll, you know, if there's a sale happening whether it's on Xbox or PSN or Steam, we think okay, what content update can we combine with that sale? Let's put the game on sale. Let's get it out there promoted, visible and in front of people so so they can dive into that new content. It's not just a sale for sale's sake. It's a sale cuz there's something new and that's really important. We really try to reinforce all of our updates as we mentioned

earlier through more media, working with influencers and social channels so people have an actionable item when it comes to what's happening with ARK. They can go see something, watch something, buy the latest content if they don't own the game yet and and then really it's it's doing all of that throughout all of development. And so that's pretty much our spiel. So thank you. Thank you everybody and uh I I want to say thank you to everybody who's played ARK and enjoyed it. We really wouldn't be here without you. Exactly. Yes. We are hiring by the way. So notice. We've got a booth. So if you are one of these people with these talents, please come talk to us. And we have time for questions if anyone has questions about anything related to ARK development or design or anything we talked about. Hey, I'm working on a fighting game called Fantasy Strike for Steam Early Access and I had actually two questions

for you. First about PlayStation 4. You don't have the early access label there because there's no such thing, but you do have it on Xbox and Steam. So how do you handle that? Do you do you frame the exact same game as finished on one platform and not the other? And the second question is that your slide at the very end showed sales from 1 million to 7 million and most of your terrific talk was about kind of that journey from 1 million to 7 million, but can you say something about how you were so incredibly successful with your first launch on early access? Sure. Can you repeat question? Okay, so the first question was about early access and PS4 and and ARK's placement on PS4. So ARK is is technically not a early access title on PS4 is that uh moniker or program doesn't explicitly exist at the present time on PS4. It is noted on the store as an unfinished game, however, or the game will be receiving lots of updates. didn't realize that. Yes. So specifically

we used to ask Sony could can we put an unfinished game on the platform as long as we're committing to finishing it and rapidly updating it and they said yes. Well, it's also worth noting we we had to actually pass full game cert which is not a requirement for game preview or Steam as you know, but for Sony's platform you have to go through cert to put even put your game on there. So Technically certified as a retail title, but planning you know, updated on a weekly basis and Sony has really good back-end systems to update the game rapidly. So it's made it very effective to operate in that way on that platform. The second question was about how we successful early on at launch. And I mean I think two things really played into that. Number one is, you know, having a a good idea that a large number of people are going to be interested in the first place. The obviously survival and dinosaurs was was effective for us. Um but secondly, it's the things we did around launch that really worked uh to our benefit. Namely, right after launch, we we were

just updating the game five times a day. I mean, I don't think anybody's updated the game their title on early access as much as we did with ARK for not just a day or a week or a month, but really the first 6 months. Now, we've obviously late more lately not slowed, but focused on larger patches released less frequently. That's in order to stabilize the technical state of the game. But very early on, our goal was to be there 24/7 to react to whatever uh issues were rising in the online game day or night. And that was difficult, very very difficult, but I think it really made an impact uh because the game the launch of any online game is going to be very rocky and faced with a lot of issues. And to the extent that we could, we tried to stay on top of them. And I think that that that helped us continue the acceleration beyond the, you know, first few hours of launch. And all the stuff we talked about with the community team, that's a very challenging time for the community team as well. They have to stay on top of the launch, talk to people, let them know what's going on, get stuff back to us. So, that's why everything we talked about, even though it was about all of the success up until now, it was really

forged in the fire of those early launch days, how we dealt with updates, how we dealt with the community. And we just kept doing it the rest of the time until now. So. Well, thanks a lot, guys. Uh hi. First of all, my favorite GDC talks are ones where developers talk frankly about their failures as well as successes. So, thank you for addressing Sheep Gate. That was really awesome. Um the question that I have is about um you guys talked a lot about um getting community feedback being a part of your scorecard system for whether or not you want to implement a new feature. So, if you are doing all this visible effort to get community feedback on a potential new feature and then you decide not to do it, how do you break that to the community? How do you go You know, that can be a tricky point of communication. So, what's your strategy? You know, that's a really good question. The The question was when we inevitably are faced with a feature that the community really wants, maybe we've even solicited feedback from them and then have to not do it because it doesn't

meet our overall criteria afterwards for inclusion into the game. How do we break that to the community and not get skinned alive? Um and the The answer is unfortunately no good answer. Uh we we have had to do that in the past and usually it just involves a very mea culpa sorry post. Uh and what we usually try to do is time it and with some sweetener. So, we have good news, bad news type situation. Good Good news, you're getting something really cool that may be unrelated. Bad news, we had to cancel this feature that you were looking forward to. And you'll get a set of players who will want your head on a spike anyway, but at least uh there'll be something to uh to look forward to as well and that's the only thing you really can do at that point. Thank you. Uh hi, hello. Uh was wondering if after the early access, you will wipe the servers or not. That might be privileged information. We We We are still considering what the

best uh options are there, whether it's to add new servers or keep the existing ones. We're leaning towards the present time adding new servers. We We certainly have resources, financial resources. So, um while having a lot of servers is is costly, uh when you look at the grand scope of player investment and new players, uh we're leaning towards adding new servers, but honestly, we're going to continue to kind of evaluate the situation as we get closer to launch. Um but that that is, you know, with any online game that goes through early access, the challenge about uh how often you wipe, if you wipe at all, and what you do when you have a bunch of new players coming in and during the inevitable uh retail influx. And so uh we don't have a perfect uh solution, but we're going to be keeping an eye on that and figuring it out as we get closer to the launch date. Thank you. Okay. Thank you. And also, please if you can add a queue system Oh, uh yeah, a queue system. appreciated. Highly requested, I think. We we uh definitely

are thinking about it. It's really technically complicated, but um yeah, it's just uh How much is the scorecard? It probably on the on the technical complication side, that would definitely be an X. But uh but definitely I I I understand your pain if you're on console, especially. Yes. Thank you. I spent uh Christmas around Christmas on cleaning dishes, pressing Trying to get into a server. every 30 seconds We we saw somebody made a mechanical device to press A button over and over again, trying to join a server. That's the that's the that was the uh ARK queue system. That was a YouTube video. Sounds like there's a solution. Why do we need to implement anything? Okay, thank you. Thanks. So, I had a question uh you touched on the end about using influencers to sort of disseminate information. And uh I know from the beginning of the uh early access you had a very robust influencer program. So, uh how did you go about spinning that up, getting uh streamers and other influencers into the fold, and uh putting out regular content about the game, and sort of building that community?

So, the question was about influencers and how we kind of build up that uh sort of community, I guess, around it. Um one thing I would say, I wasn't just talking about influencers in the terms of, you know, voicing the game and, you know, we're trying to get them to get the word out. Influencers influencers in the sense of content creators, like on YouTube, uh Twitch, people that are creating content for the game, or they have their own fan bases, and we really try to reinforce their popularity, give them cool stuff uh to work with uh report on from the games development. Occasionally we've done like early look at something but we tend not to work that way cuz we're always so close to launch but it's important with the influencers I think to reinforce the good work they are doing. If we have a video that we think is good from a content creator will retweet it. We'll give props to people that are creating content for the game. If it's you know music video or it's funny comic that somebody's made. Those type of influencers are not like direct oh I'm playing the game watch me on Twitch.

Those are influencers that are spreading the word about the game and I think that's kind of the answer for us is reinforcing that relationship by being a an amplifier to the stuff people are doing about our game outside of us. I think we're almost done. Was that the no more questions signal? Thank you everybody. So, we'll hang around a little bit.