How can developers successfully launch games on Roblox?

In my recent conversation with Zach Letter, founder of Wonder Works, we discussed just that- the business and marketing of Roblox games. Wonder Works is an influencer studio that have created hits such as Overlook Bay and Traitor! Those two games alone have had over 200 million plays alone.

Let’s jump in. Below is an edited transcript of the full interview.


To start, I love to hear about how you got started with Wonder Works.

I’ve had a very untraditional background when it comes to gaming. When I was a small child, I loved video games. My first gaming memory is playing Conan the Barbarian on the NES. But when I came to being an adult and having a career, I was fortunate enough to find YouTube really early. I started producing gaming videos and content in 2007 and turned that into a career for the better part of a decade. I amassed a few million subscribers and did over a billion views across all my channels.

Then I met my wife and in 2018, she wanted to start her influencer journey. She was doing a few things prior to 2018, but she started looking at Roblox and said, “wow, this platform’s huge and it seems really diverse and I think I can build something with the IP that already exists there”. So she built a massive audience in just two short years. She has 10 million+ followers across all her social platforms. She’s done over a billion views as well.

During her rise to success, I noticed she would have like 70k people watching a live stream. And I’m like, wow, you can command a lot of people to go do good, check things out. You can introduce them to new IPs and franchises. I’m like, maybe we should make our own game and it’s something we could do in partnership with your audience and they can be part of the process.

And so in 2019, we founded Wonder Works with that idea in mind, using social media to build a presence for a studio where the studio itself is an influencer in a way.


I’ve never heard anyone describe a game studio as an influencer. Can you explain that?

Wonder Works has a few unique approaches to user acquisition. I’ll start with how we acquired users on our first IP, Overlook Bay. I designed the experience from the ground up and then a small team helped develop it. Just a team of two- one programmer and one environmental artist. We were kind of like guys in the garage trying to make our first product and see if people enjoyed it.

When we were making it, we knew it was going to take us seven or eight months to produce. And we thought, ok, how are we going to market this game? Obviously, the low-hanging fruit was to use Megan’s brand and push as many people to it as we can.

It’s about utilizing viral design, viral tactics to acquire users.

But we were really new to making games. We love video games, but we didn’t know like what people really wanted on Roblox. We didn’t want to put Megan on the line, pushing millions of people to this game if it wasn’t ready.

We decided to make a virtual influencer or a VTuber called Honey the Unicorn. And this was a character that existed in Megan’s universe, and only appeared in Megan’s videos. Spoiler, I was Honey the Unicorn.

We decided to make an animated face cam that we would put through Adobe, and we were going to give it a unique voice, a bedroom, a personality, and a webcam that it would then record gaming videos on. And our plan was that it would play games that were of the same genre as Overlook Bay on Roblox.

Every audience member that they acquired would potentially be a player of Overlook Bay. And in just a few short months, they gained nearly three quarters of a million subscribers and we were doing millions of views every single month. And within two months of Overlook’s release, we gave Honey the Unicorn early access.

He was doing demos and playthroughs, getting audience feedback and he became part of our development team. Getting all these people watching all these videos and giving their feedback was so helpful to the design process. Because of this, we had over 1 million favorites on Overlook Bay before its release.

There was so much attention across the entire Roblox community that we decided to release the game in paid access. And we sold 1.2 million copies at 25 cents each in two weeks. Within 4 weeks of release, we dropped it to free-to-play and had over 90k CCU and 26 million unique users.

So that’s how we think of us as influencers. It’s about utilizing viral design, viral tactics to acquire users. We never pay to acquire a user at Wonder Works. That’s something we really pride ourselves on.

We figure out how we can make content that’s organic enough to catch an influencer’s eye. How can our icon and our title speak to an influence when they’re scrolling through a page of hundreds of games.

How else do you catch an YouTuber’s attention with your games?

We have a network of hybrid casual experiences that have somewhat replayability. They’re not totally hypercasual, but we release one to two of those experiences every single week. To date, we have released six of them since late February.

We’ve done nearly 100M visits on those, but over 100M views off-platform as well across those six titles. And one has nearly 50M on its own. And that’s not counting TikTok or YouTube Shorts. If we count those, it’s tens if not hundreds of millions more.

It’s also about having an antagonist on the icon that’s a little bit disturbing, maybe not super appealing

Let’s say we’re building a platformer. It needs to have a narrative and story to really appeal to YouTubers. You need to set them up with something.

The dilemma of a YouTuber is having to produce daily content. They’ll say things like, “What am I doing today? I need something. I’ve been playing the same game for X months or X amount of years. Is there anything new that could help me out?”

Our strategy with the HyperCatz network is we’re going to release two games a week for you. That should be a minimum of two videos out of your seven. And when we service LiveOps to these games as we have more in the network, we’re hoping that we can take up your full seven day calendar.

It’s also about having an antagonist on the icon that’s a little bit disturbing, maybe not super appealing. We’ve had some success with that, for example like Escape Your Evil Stepmom. We made her look a little bit ugly, a little bit robotic, and a little bit dead in the eyes. And kids were like, okay I’m scared of her but I wanna see what’s going on.

We also really take a Mr. Beast approach to how we design these games. It’s about having a story that you can have a sense of completion in a session. And also, a lot of content nowadays don’t have you stay on one thing for too long. If you watch Mr. Beast’s videos, there’s a cut every three seconds, if not less. He’s never keeping your gaze on something for more than just a few seconds. And when we were designing our platformers, a lot of these 10 minute experiences have like 45 stages.

It’s like the video basically make themselves. We’ve noticed YouTubers doing literally a playthrough of these experiences and then not editing them and then getting three million views on a video.


What are your thoughts on the 29% developer cut on Roblox?

One thing I always say to their defense with their take is that we never have to think about backend. We never have to think about servers. Our game could have one player or one million CCUs and our cost structure wouldn’t change a penny. So I think when it comes to developing games with broad appeal, it’s going to have a lot of users.

We don’t have to think about things like:

  • Are we going to have our servers crashing?
  • Are we going to have to backup data?
  • What are we doing with the players’ data?

Roblox bears all of that responsibility for us. And I think that’s a decent trade.


What’s the roadmap ahead for the Wonder Works team?

Our goal this year is to have over 100 million monthly active users across our HyperCatz network and our flagship IPs. We’ve moving towards that. We’re nearly 40 million monthly actives right now. And we’re going to continue to expand to other platforms such as Fortnite. We’re actually releasing three Fortnite experiences this summer, along with probably 15-20 Roblox experiences as well.

Our ultimate goal at Wonder Works is we want to create IP that could be the next Pokemon, to create that forever game. Create something that can stand the test of time, go to franchise, go to direct-to-consumer retail, and series too on streaming platforms and services.



We cover so much more, including the Roblox discovery algorithm and revenue numbers. If you want to hear more, check out my full conversation with Zach Letter.

If you enjoyed this article, tweet at me @jackzliu and let me know your thoughts!

Get more insights like this on the business and marketing of the metaverse delivered to your inbox. We’ll never spam you, ever.