“We’ve learned a lot, but our mission to make great horse games hasn’t changed” – Reflecting on Equestrian The Game with Kavalri Games’ Molly Ericson

Equestrian the Game is a mobile game about owning, training and breeding horses. It was released globally in Spring 2022 after months of regional early access, and after being announced and widely followed as far back as 2017. 

ETG stands out as a significant milestone in the horse game genre: a horse game made outside of the traditional game publishing industry, helmed by people with a burning passion for the subject matter. A game for equestrians, by equestrians. By now, (and how cool is it that we’ve come so far?!) ETG is far from the only horsie passion project anymore, but it was the first of its kind with a comparable scope and success. 

In celebration of the game’s Pinto update a few months ago, Kavalri Games’ co-founder and CEO Molly Ericson sat down for a call with me to discuss some critical learnings about making free-to-play games and how to handle players and their feedback, about what’s left in store for ETG and what we can expect from Kavalri games in the future.

Pintos and Prosperity

Molly Ericson is the Co-Founder of Kavalri Games and has recently shifted her responsibilities from Creative Director to CEO.

I’ve published Developer Insights from Kavalri on two earlier occasions – from initial intentions back in 2019 to launch excitement in 2022. Despite these exchanges – and despite attending the same industry events on occasion – our call for this article in May 2024 is the first time Molly and I actually talk. So how is the ETG Team doing?

“We’re doing good!” says Molly. “Since the launch in 2022, it’s been all about building the features we had always seen before us, working on seasons and content updates – we’ve been very busy.”

Not only the game has grown, but the team behind it: Kavalri now consists of twenty people, working in a hybrid model in Stockholm, Sweden. “Obviously it’s been challenging times for the industry as a whole,” Molly adds, referring to the wave of studio closures and mass layoffs that have taken place in the past months. “That’s been a bit of a rollercoaster, not going to lie. But in terms of just Equestrian, we’ve been working on things we’ve always wanted to do, like the Pinto update that released recently. That is something players have been requesting since day one when we were building our first prototypes, and it’s been super amazing and exciting to finally deliver that.”

Horse breeding is a key feature in Equestrian The Game, with many players breeding for both stats and personal breeding goals like specific colors, body types or sizes. The breeding feature so far has been limited to solid colors, but it has always used realistic genetics as a basis, pleasing the sort of horse nerd who knows you have to pair a cremello and a bay to get a guaranteed buckskin (it’s me, I’m that sort of horse nerd). 

Within Kavalri, a three person team has been working exclusively on the pinto feature, under the leadership of developer Jasmine Öhlin. Jasmine came to Kavalri with a lot of pre-existing knowledge and passion for animal genetics and was thrilled to apply that expertise to this particular implementation. 

“Jasmine really deserves a special mention for adding this whole new dimension to the breeding system,” Molly says. “She’s been consulting experts in the field and looking at gene-tested horses to analyse and explore how different white patterns interact with each other.”

Molly explains that they could have gone for a simpler version, something quicker to implement with more randomization. “But we wanted to do this properly. We know that many of our players are really into horse genetics and will appreciate the level of detail.”

Pinto horses can be bought from the shop…

…or created in the game’s horse customization.

And still, some simplifying was necessary, considering that there are dozens of specific white patterns in horses and the science of it is constantly evolving. Kavalri uses Tobiano, Frame Overo and Tovero as their white pattern types, but (T)overo can imply a combination of different possible genes including Sabino and varieties of Splashed White in addition to the Frame Overo gene itself. If that’s a rabbit hole you’d like to get lost in, I can only recommend joining the Horse Color Genetics Facebook group. Defying Facebook tradition, that group is a legitimately incredible and reliable resource.  

A demonstration of how shader effects add visual diversity to the underlying masks.

“We had to figure out how to add visual diversity – ensuring that pintos wouldn’t all look the same – while optimizing and not bloating the file size with hundreds of individual pattern files,” Molly explains. The solution they settled on is a clever combination of predefined textures and modification on the go: “As a base, we’re using thirteen different masks, and then we adjust these a lot with shader effects, so we can control how much of the mask is actually visible.” 

Another consideration for the feature was keeping it accessible and intuitive, despite the complexity of horse color genetics as a topic. “We don’t want to overwhelm new or casual players who just want a pretty horse.” 

After launching the pinto update, Kavalri immediately observed a lot of engagement with it on social media, including players showing off particular combinations that they could have only gotten through careful breeding. “It’s been super exciting to see,” Molly adds. “I’m really proud that we’ve gotten to this point where we have the team with all the skills and resources to needed to deliver this feature.”

It’s a sign of maturing for Kavalri, to have a multi-disciplinary team to set on features like this – they’ve come a long way from the situation around the game’s launch, where Molly used to still handle all character art by herself.

Community, Feedback and Boundaries

Another aspect of the game’s creation and maintenance that Molly used to be handling herself and has since stepped back from is its community management. Kavalri had set up a Discord Server for its players, initially intended to be a space for a few dozen select beta playtesters to give in depth feedback. During that time, Molly talked to testers one-on-one, was available by direct messages, and even hopped on calls sometimes to clarify specific bits of feedback. “It was a very personal experience,” Molly says, “and it took a lot of time: I would spend full days doing that, which was valuable and needed at the time, but led to some players getting used to having a direct line to me.” 

Around the game’s release, the server was opened up to the public, and quickly grew from about 30 members to over 9000 within weeks. The same kind of availability and personal interaction became impossible to maintain due to sheer volume. 

“We had hired a Community Manager by then,” Molly explains, ”but we also had the challenge that as I stepped back from that role, some players felt as though I was abandoning them, because I was still the one with the knowledge of the game design, the decisions being made – which a new Community Manager doesn’t necessarily have.”

These circumstances led to disappointment and to a very negative atmosphere in the server, and eventually the closure of the Discord community, as it simply became unmanageable and unsalvageable. 

“In Retrospect, it was great to have the server for the original purpose, the beta community,” Molly reflects, “but for a larger, public community we should have managed it in a different way. You can’t handle nine thousand people in the same way you handle thirty people.”

The topic isn’t a pleasant one to discuss, even two years after the server’s closing, that much is plain to see in Molly’s expression as we talk. “It was a harsh experience for me personally,” she says. “Because it’s never fun when people online get so angry and upset at you.”

There is value in the lessons learned however: “For a long time I set this expectation on myself that I have to keep the community informed, that I have to be there for them no matter what time of day. Since I was still doing art for seasons passes at the same time, I would do that work on weekends and evenings, feeling like I owed that to the players.”

The server blowing up in negativity taught Molly that this wasn’t a reasonable expectation to have for herself. “I realized that players don’t care about me, they won’t know if I get burned out – they don’t know me, and I don’t know them.”

She says it was a wake-up call that she cannot be that personally involved, that she had to step away. Her experiences reflect many that I myself have had over the years: Especially when doing it all out of passion as we are, it’s easy to get carried away on the notion of wanting to please everyone with an interest in your niche, your game, your community. I myself know that I’ve hardened over time, that I do a better job nowadays of protecting myself from having to bend over backwards for the people you’re hosting in online spaces. 

Molly wistfully agrees: “It’s so easy to get swept away when you want to make everyone happy and people are mad at you. Even after I stopped managing ETG’s social media I would get toxic comments on my personal Instagram and try to engage with people, trying to be super nice and take the time to explain things to them.. Sometimes it worked, and people would apologize and realize I’m an actual person,” she laughs, “but then you’ve convinced one person and there’s millions more out there.” 

As someone with a chronic case of ‘I can fix this with my words if I try hard enough’, I deeply relate to that tendency, and tell Molly as much. Words can have that power, but in a community management role you need to respect your own time enough to make smart choices about where you invest this sort of effort and energy. 

“Yeah, you have to set those boundaries and decide whether individual discussions are worth your time,” Molly agrees.

She connects that balancing act to decisions made about the game itself: “If you focus too much on what the community wants, or you’re afraid to make changes for fear of community backlash, your hands are tied as a developer.”

One crucial aspect to remember here – and one that vocal fans and critics are often unaware of – is that any game’s online community only accounts for a small fraction of the actual player base, especially in a free-to-play game like ETG. “That’s also been an important learning for me,” Molly adds, “to see the bigger picture, and not just focus on the players who are actively communicating.”

It’s general free-to-play market wisdom that only a small fraction of total players spend any money on the game at all. For Equestrian, about 1.5%-2% of players spend, with spikes noticeable when new seasons or horses release. Those numbers are in the ballpark of industry average. When asking developers to “listen to players”, one has to keep in mind that the venn diagram of paying users and vocal users is never a simple circle.  

All of that does not mean that Kavalri has stopped paying attention to what players ask for, but they’ve changed the methods by which players can reach them. Nowadays, Kavalri has a Google Form where players can submit ideas and suggestions, and their Community Manager compiles the gathered feedback into quantifiable, easily digestible data for the Game Designers and Producers. A format that saves time and nerves in the team, but still gives players an avenue for input. 

On Google Play alone, ETG has been downloaded over one million times.

Monetization and Mobile Audiences

Despite these growing pains, Equestrian The Game has been a success:  Kavalri received several funding rounds from Amplifier Game Invest and Coffee Stain Studios initially, but ETG income has since sustained the studio’s growth, despite the uneasy times of 2023 and 2024 amidst layoffs and closures all throughout the industry. 

“The game was monetizing… somewhat,” Molly laughs. “It sustained us at a certain size, but at the same time you want to be able to expand a little bit, to also get more stability as a studio. We want to run a profitable business and save costs, but also be an a competitive workplace that attracts good talent.” 

That goal has been in the background of ETG’s development throughout the past few years. 

“I see Equestrian as a game with so much potential,” Molly says, “There are a lot of players, a lot of people downloading it spontaneously and enjoying it. But one of the challenges for us has been that when we started designing the game, we didn’t really design it as a mobile free-to-play game – we just wanted to make a good horse game.” She grins and adds: “That was the dream! We got funding, we get to do this, let’s make a good game.”

“We really tried our best to do that,” Molly continues, more sober, “but we’re in the situation were we just launched it first and now we’re trying to keep developing it into a successful free-to-play product.”

This proves challenging with a game that’s already live: Players quickly get used to mechanics and don’t always appreciate changes – especially ones that are plainly intended to increase revenue. When Kavalri started out, they were a very young team of first-time developers, all without experience in creating this kind of product. How precisely the game would be monetized was far from the first thing on their minds, and they’ve been playing catch-up in various ways since. One example of that is that the option to watch ads in order to get certain in-game benefits or premium currency was only added last fall, which was not so much a conscious omission at launch, but simply not something that was initially considered. 

The game now features incentivized ads, where player watch an ad and get premium currency in return.

“There was just so much we didn’t have in place in the beginning,” Molly adds, “and that has definitely harmed the profitability of the product. We’re still here, we’re a twenty person studio now, so it’s somewhat successful, but we all know that there’s so much we could be doing better!” 

Considering these struggles – including the difficulty of how to communicate changes to the monetization to an enthusiastic playerbase – I ask Molly if Kavalri would choose to release ETG as a premium title on PC instead, if they had to make that choice again. 

“I still think it was the right move for us to go to mobile first ,” Molly ponders, citing the ubiquity of mobile devices among the target audience. Even so, she doesn’t see Kavalri as exclusively a “mobile studio” even now. Releasing a game – be it ETG or any future title – on PC is still somewhere on the horizon, but comes with its own many considerations for market, reach and audience. “Our vision is still to expand beyond mobile.”

A/B Testing and Optimization

The way Kavalri has been going about optimizing their monetization and turning ETG into a commercially viable product has not always been welcomed by the community. Price increases or adding paywalls to previously free features tend to provoke negative reactions in long-term players. In discussing the monetization decisions of games, players are often quick to speculate along the lines of “well obviously SO MANY people are annoyed by this change that the game isn’t actually making any more money” – while once again forgetting that the vocal community members only ever represent a tiny portion of actual paying users. 

Examples of ETG’s monetization: horse shop…

…and season pass

Still, it is absolutely possible for monetization changes to do more harm than good: for example, to lead to an increase in short term revenue, but an averse effect on other crucial factors like initial conversion or long-term user retention. To navigate these considerations, developers use A/B Testing. That means part of the user base is shown one version (of pricing, of user interface, of wording…) and another part of the users gets a different version. Key indicators like conversion, retention, and overall spending are then tracked across both groups in order to find out which version performs better under consideration of all relevant factors. 

Like many Free-to-Play game developers, Kavalri employs extensive A/B Testing when tweaking their monetization. That doesn’t mean all questions about which version is better for business in the long term are easy to answer though: “One challenge is that we face is that we want to be able to make changes quickly,” Molly says on the subject, “but at the same time we need to gather enough data – ideally seven or eight weeks. You also might want to test several things at once, and not restrict yourself too much in making changes to the game while tests are running.”

Molly adds that there are external factors influencing their A/B Tests that are hard to account for, such as seasonality – both real life seasons as well as in-game events. “We’re constantly trying to find that balance of being data-driven, but also acknowledging that we can’t always be one hundred percent certain when choosing this thing or that.”

On that note, I dig down more specifically, due to some community speculation I’ve seen about how recent changes to make the monetization more aggressive must surely have resulted in fewer players spending at all.

 “It’s generally been tricky to for us to get more than a specific ratio of players to become payers,” Molly explains, adding that the rate of paying users in the game has been very constant around 1.5-2%. This has stayed the case throughout this year’s monetization changes, disproving the aforementioned claim. 

“This too is a balancing act,” says Molly. “If you have a purchase that’s very very cheap, that might look good in your dashboard because that conversion rate is suddenly 5% instead, but the revenue of that is so low for it to still not be worth it.”

These learnings, that lowering or increasing prices may not actually affect the percentage of paying users, or that more people spending can turn out to be not worth it if what they spend is little enough, is something that’s often quite unintuitive to players, and something that many mobile game companies don’t bother to give a lot of insight into outside of exclusive industry events. 

Molly summarizes that if they weren’t seeing positive results from such changes – or if the costs to player retention and conversion were too great – the changes would be reverted. 

Although we didn’t talk about it in detail in our call, I did send a follow up email to ask ask Molly about the choice to use AI-generated images in ad campaigns on e.g. Facebook, since I’ve seen various players question that choice and especially its ethics, considering that AI image generation models are generally trained without the consent of the artists whose work they are based on.

AI-generated ad on Facebook. Note that both horses in the image have extra limbs. Source

AI-generated Ad on Instagram. Note the “horse” in the background on the right for an obvious tell. Source

“We've been consistently testing different types of campaigns,” Molly replies, explaining that these campaigns compare handcrafted art, in-game footage, concept videos and stills as well as AI-generated imagery. “I understand the sentiment”, she says, “at the same time, the AI generated stills have been performing great, which is why they're consistently used.”

Molly adds that User Acquisition is a huge factor in the success of any Free to Play game, and that the AI-generated images have been a big boost for ETG in that regard. “That said, AI is a topic that's both exciting, sensitive and thought-provoking. It's an extremely powerful tool that's used by us daily in various ways. The best practices and how-to is something that's subject to internal discussions within our studio and our stakeholders right now. We still have some ways to go in terms of defining a framework for how we approach the topics of incorporating AI in our work going forward.”

ETG and Beyond: Kavalri’s Future

Asked about the studio’s plans for the future, Molly reveals that Kavalri is planning to make more games rather than forever expand their first project. “There is something in the works that will become our next game,” she teases, then laughs. “It’s in a very early stage, so there’s not that much to say about it, but we’re very excited to be making a new game – something Kavalri hasn’t done in a while.”

Kavalri Team Photo

For now, all we know is that said next project will once again prominently feature horses, and be developed for mobile devices. “We absolutely want to make a smaller project – to take a bit of time to release something small to give the team that win.” 

What does that mean for Equestrian and its planned features? 

“We are approaching a point where we’ll consider ETG ‘feature complete’,” Molly tells The Mane Quest exclusively, aware that this will come as a surprise to long-term players. “We had very many features planned initially,” Molly says, cringing slightly at what I can assume is past naiveté at the feasibility of such a long feature list. “By now, we are having important discussions as to what we actually can do on this project, and what may have to be removed from the roadmap even though we’ve always wanted to do it.” 

I call back to how I constantly tell first time developers to keep their project scopes as small as possible, because before having finished a game, many of us are actually absolutely awful at estimating what is a reasonable amount of features to promise on what sort of a timeline. Something that Kavalri definitely also has had to come to terms with and learn from. 

“The game has been in development for about seven years now,” says Molly, and explains that this long lifetime means that some of the initially built systems absolutely do not lend themselves to being updated. The game’s tutorial is one of those things Molly’s been wanting to update for a long time. “Breeding isn’t even in the tutorial, because the tutorial breaks as soon as you touch it.” I laugh at that, knowing it’s not a joke, because well, ‘don’t touch this part of the game or it completely breaks’ is something that most game devs will have experienced at one point or another.

The result is that certain changes look like they’d be very simple to the outside observer, while those familiar with the project’s code know that even a slight adjustment would take a complete technical overhaul. “For those cases, we know it will not be worth it, because it would take two devs six months to do. In that time they could make a whole new horse game! We have learned so much about how to make mobile games that that would be a much less painful experience.”  

That there comes a time when projects are concluded and studios want to move on is something I see players have very little understanding for – especially those very vocal players who refuse to make any difference between games being finished or games being abandoned. At some point, games are done and that is fine.

The issue admittedly becomes more complicated if there’s genuine learning on behalf of first-time devs like Kavalri where they start out fully confident that a specific feature will be added eventually, but then need to come around and say that actually, no, something that was initially promised won’t be feasible after all – as Kavalri had to do in the meantime with their announcement that the long-planned Dressage feature has been scratched from the roadmap. The kind of player who’s loud in comment sections rarely reacts with grace and kindness to such news – and I’m asking my readers to consider this when voicing their own reactions to this article. 

“There’s a point at which further polishing something so flawed is simply not the best choice anymore,” Molly summarizes, agreeing with my recurring takeaway that more devs should plan their first several projects as tiny as possible. “I love this game more than anyone, but even I am starting to get ready to move on, to make something new! Even though it is very scary to do so as a founder since our entire business was built on Equestrian as a starting point.”

With all this talk of Equestrian’s development approaching an end, Molly still tells me that there are some exciting features still in the pipeline: Most importantly, Stable Customization and Cross Country riding, which have been officially announced in the meantime. “Those are two pretty major features that were a long time coming, that I think players will be very happy about!”

“There’s been so much uncertainty within the games industry in the past year, but I can say that we’re in a good place, we’re in the best shape we’ve ever been with a very passionate team. That’s part of why I’m so excited about our upcoming projects: We have really stepped up our game and will be able to deliver something cool, better and faster. Our mission to make great horse games hasn’t changed.”

I for one, am quite excited to see what Kavalri announces next, and how they’ll be able to apply their myriad learnings on how to make good horse games to future projects with more manageable scope.