Our coach has a very effective style of facilitation and introspection injecting provocative points into the contexts that leade to valuable extensions to the discussion.

Alan Young, InRule, Chicago, IL

Validate, Replicate, Formulate, Automate. FRVR’s Formula For Startup Success

Interview with Brian Meidell (CEO) & Chris Benjaminsen (Founder) of FRVR.

Validate, Replicate, Formulate, Automate. FRVR’s Formula For Startup Success


Show Notes

Do you play the games you like best—or the ones that have been marketed to you the hardest? This is a huge problem in the gaming world. FRVR is hoping to fix a broken system by helping gamers play the games they’ll enjoy the most while also helping game developers reach their audience better.

FRVR co-founders Brian Meidell and Chris Benjaminsen joined Roland Siebelink on this week’s episode of the Silicon Valley Momentum Podcast. In addition to discussing how FRVR is moving things beyond the traditional app store, the two entrepreneurs shared a variety of thoughts on the gaming industry and their keys to startup success.

Transcript

Roland Siebelink: Hello and welcome to the Silicon Valley Momentum Podcast. I’m so excited today because joining us, from I believe Malta, are Brian Meidell, CEO, and Chris Benjaminsen, the founder of FRVR, is that pronounced as “forever,” guys?

Brian Meidell: It is F-R-V-R. If you absolutely want to pronounce it, you would probably say forever. I don’t know if this is useful, but we are, in fact, joining from Lisbon and London.

Roland Siebelink: See, I got it wrong already. This world. We have no idea anymore where people are, right?

Brian Meidell: We are a very multinational bunch, so that’s perfectly understandable.

Roland Siebelink: FRVR, what does the company do? What difference does it make in the world?

Brian Meidell: FRVR makes games and brings the right games to the right user wherever they are. And that sounds very simple. It’s not currently what’s happening in the world.

Roland Siebelink: Okay. What’s the difference from other companies that make games? Because there are quite a few out there, right?

Brian Meidell: There’s quite a few. We are mostly focused on the world beyond the app stores. Most of the game industry currently would be on mobile app stores, I think. And today people on mobile devices is basically most of what we consider computer users and gamers. But on mobile, the distribution model is just fundamentally broken.

Generally, people end up playing the game that was advertised the heaviest to them. That makes it so that the people who make games generally have a real hard time reaching their audience because they can’t compete with a hundred million dollars marketing budgets. And the people who play games generally get to play the games that monetize them the most aggressively. And we don’t feel like that’s an ideal situation for the players or the developers. We are currently trying to solve that.

Roland Siebelink: Does that mean, Chris, that, in a way the game companies become more marketing machines than actually good game developers that you don’t really compete or that companies typically don’t really compete in the quality of the game anymore?

Chris Benjaminsen: Yeah. Yeah, absolutely. Making a good game today is probably less than 10% of actually being a successful game developer. Of course, you have people who are lucky who stand out. The people you hear about, where it’s a single developer who just happened to make that game that was good enough, I think YouTube picking it up. But that is an unnatural state of affair. That is not the story of the average developer.

What we’re trying to do is we’re trying to get the games in the hands of the users without them necessarily having to go through this inherently abusive process of user acquisition. And what we found by doing so is that the consumers don’t inherently have a preference for technology. As long as they get a good game that they can enjoy they don’t typically even know or care about how they got that good game. And that has opened up an opportunity for distributing games in different ways.

When I was really young, the way we would do it is we would find a user on the Internet through a browser, a better version of the game. And then we would up-sell them to playing the mobile app version. But as the market has matured and as mobile phone has become more powerful, that need to move people into a mobile app has gone away because now we can provide sometimes even a better experience outside of the app store than inside the app store.

And that might sound counterproductive, right? If you look at the average consumer today, they have much more than a hundred percent of the time taken up by entertainment. People are playing games while watching shows or whatever. People already have something to do. There’s no minute in the day where a consumer is not entertained, right?

Roland Siebelink: The end of boredom in a sense, right?

Chris Benjaminsen: Yeah. Yeah. There’s no boredom, right? We’re all fighting for the same eyeball minutes now. And that means for you to provide a product that a user wants to engage with, or is likely to engage with, you’re much better off trying to address the user where they’ve already decided they want to be. If the user really likes TikTok, allow them to play your game on TikTok. Rather than necessarily trying to do the thing which happens now, which is drag the user away from TikTok through the app store to play your game on the mobile app.

Roland Siebelink: Brian, does that mean that part of the strategy is to be embedded in several platforms to, as Chris was saying, really be where the user already is and not force them into a friction process of having to sign up, move into an app store, things like that?

Brian Meidell: Yep. You totally got it. We let the user decide. And I think today, users don’t generally - they don’t go to the specific place to consume the specific content anymore. They’re used to these rivers of transient content flying past them and newsfeeds and messengers. Basically, they can decide to engage with not just games but also music and movies pretty much anywhere they are. And we just feel like this is the latest media format to follow that pattern.

Roland Siebelink: How many games do you have right now? And how’s it looking in terms of the distribution of users? Is there one game really having the bulk of all your traffic or is it more of a distribution that is also maybe shifting over time?

Brian Meidell: It actually shifts over time. But in general, I think all game portfolios will always have some sort of power law distribution. There will always be a few number of titles that have a relatively high percentage of the traffic. We’ve seen over time that we’ve gotten better and better at compressing that field.

It’s a little bit special for us because we don’t just operate in a single market. We operate in many, many markets on top of each other. We tend to have these wild growth spikes in one market or another market, which will make whatever I say now true right now but probably untrue by the end of the conversation. It certainly is getting better and better over time to the point where quite a few titles are seeing some really significant traffic.

Roland Siebelink: Okay. What I think a lot of other founders could possibly learn from here is how do you manage so much complexity? Many other startups I talk to, scale-ups even, are managing one product in maybe one geography and thinking about more geographies. How have you guys been able to manage that complexity with multiple products in your portfolio and then also in multiple markets? What does that mean for how you set up your company?

Chris Benjaminsen: Basically, FRVR was built exactly for its use case from Day One. It was always the case that we wanted a build once and distribute everywhere model. One thing we, for instance, we do with every single developer we work with, they all work in the same, technical framework. There’s no exceptions to that. There’s only one way to build a game that we can publish. And that’s through our game engine and through our infrastructure.

And we also generally only have one deal for developers. We have the right to actually do this distribution globally on any platform. And it is a big technical challenge to support that. But because we can amortize that cost across the entire catalog, it is possible to do that in a very economically beneficial set up.

Brian Meidell: I think it’s also worth mentioning that from Day One, we built everything as a programmable piece of tech. Everything from getting a list of our games to which phase of development they’re in, all that stuff is captured programmatically, which means that from Day One, we could treat the entire catalog as something we could manipulate with more programming. We’re both very - we’ve both been engineers for decades. By capturing that complexity in a way that makes us able to deal with it with machines, it allows us to let the computers do all the boring work.

Roland Siebelink: Very good. As you mentioned, your history of having been engineers for many decades. Maybe we can delve into that a little bit. How did FRVR come together? Where did you guys meet? In what way did you decide to launch the company and what has changed since you started the company?

Chris Benjaminsen: The hyper-condensed founding story is that I had a corporate job in America and I was tired of having a corporate job in America. I wanted to go build - I called it a hobby startup. I blogged about this aggressively when I did it. The idea was to get to making 300,000 dollars a year because you can live most places for that while doing as little work as possible. This was based on the idea that if you make games that are good enough that people will engage with them, people would play those games forever. And then the only thing you need to do to be successful - I say “only” in quotation here - is it’s be good at distribution.

While I was working in this job, I managed to build a solitaire game, like Windows solitaire, the solitaire that everybody knows, and grew that to $150 a day in revenue. If you can do that eight times, you’re done. You’re already there.

I’d been working on this next game. I wanted to try to see if I could build a game that was not a known game but a newer mechanic. And it turns out if you put a pretty good new game, something that actually had a little bit of innovation, into a machine that can pump a solitaire game to $150 a day, what you’re making is $3,000 a day. That was the reality I was in a month after quitting my previous job. I’d been working with this for half a year. Quit my job, one month later, I was making $3,000 a day. I was like, “Oh, that was easy.”

And I did the most Silicon Valley thing you could, actually, when I quit my job. I sold all my stuff and started traveling. But it was very apparent then that that was more than just a hobby startup.

Roland Siebelink: A big opportunity that you hit on.

Chris Benjaminsen: Yeah, huge opportunity. There was a value difference in the market that could be arbitraged. And there was a willingness from consumers to play these games. The realization that people did not care where they play the game, only care to get a good game was the big insight. I started thinking about doing publishing and I started thinking about building a real company.

And I was talking to everybody I know about this. There was no secret here. There was no secret sauce. There’s nothing I had to keep secret. I just talked to everybody about what I was doing to try to get a bunch of feedback. And Brian was one of the people I’ve known for quite a while. I used to rent office space from him. And Brian had managed to extract himself from the company he was building before. He sold some of it off and done quite well for himself as well. Brian is good at all this stuff I’m bad at.

Roland Siebelink: Talk to me about that. What are you bad at, Chris?

Chris Benjaminsen: I am incredibly bad at structure and repetition and building teams of people to go do things. I’m a decent generalist and I’m pretty good at growth hacking. I have people working for me now, but I probably shouldn’t be a boss. And I most definitely shouldn’t be a CEO. And it’s this match made in heaven. Brian joined as a late-joining co-founder and we fairly quickly grew from two people to 14 people based on this initial success.

Roland Siebelink: Yeah. Yeah. That’s awesome.

Brian Meidell: And we started unpacking the strategy to realize it was a much larger market opportunity. It was very clear that there was a paradigm shift happening where the 10 years of ironclad rule of the app stores was starting to show some cracks. And it’s interesting, for all the time that we’ve worked together, we’ve never pivoted or changed the direction of the strategy. But it has become both more focused and execution and way more expansive in ambition. It’s interesting to see how it’s continued. The red thread is very clear all the way back. But the ambition has just grown insanely since we started.

Roland Siebelink: Focusing on the crack in the app stores, it sounds like, and then you see more and more of a crack coming through. Is that right?

Brian Meidell: Yeah. Actually, when we first started walking around at conferences, talking about this stuff to people, it was a pretty fringy lunatic idea that there was something beyond the app store. You’d have people that have grown up their entire professional careers inside of the reign of the app stores. They consider the market dynamics of the app stores their laws of physics. They’re immutable. But we’re old farts, so we’ve been around the previous 17 game distribution paradigms. And we could recognize the winds of change when we started seeing them.

Roland Siebelink: Brian, coming back to the co-founder dynamic, what do you appreciate in Chris that maybe you’re not the strongest at?

Brian Meidell: You’re great at - you ask the questions that I want to answer. I wanted to jump in before and say I think that the thing I really appreciate about Chris and how we complement each other super well is the fact that he may be bad at structure, but he is the fastest person I’ve ever seen at trying stuff out. He is the most prolific prototyper and his prototypes are solid. When you get things into the hands of customers, it feels good. It might be made with duct tape and baling wire behind the scenes, but the customer will never feel that.

The cool thing about that is that, I think as an engineer, more typical engineer that I am, you have a tendency to over-invest in tech. You tell yourself that it has to be good before it can be used. And Chris has just so profoundly shown me that that’s not - the product has to feel good. But you can make do with 10% of what you thought you had to do before it actually makes sense and can get it into people’s hands.

Chris can write fantastic code when he wants to. But the fact that he is super fast at testing stuff out, it very quickly gave us this dynamic where he found stuff that works and then I made it solid, basically scalable. And that was how the entire tech foundation of the company came to be and why we ended up being able to capture so much complexity was because I wouldn’t over invest in things that weren’t already proven and Chris wouldn’t be bogged down with having to polish everything up to a place where we could stack something heavy on top.

Roland Siebelink: Yeah. It just reminds me of, I believe it’s Matt Mochary, who in The Great CEO Within, wrote about being conscious of the difference between prototype code and production code. And that many startups fail at that because in the beginning, everything is a prototype and then two years later, they’re bogged down in bugs and everything needs to be very solid. And then they wonder why the product doesn’t deliver any features anymore.

How do you guys think about that balancing of prototype code and production code? And would you actually, as an ongoing question, as a supplementary question, would you actually recommend that people think of these as separate buckets of the product engineering departments?

Brian Meidell: We ended up finding a methodology that we call the four ates. Which is a smart way of saying - a clever way of saying we usually separate stuff into four stages and they’re of wildly different length. But it really helped communicate how we do this stuff.

The four ates: validate, replicate, formulate, and automate. And we start by validating something as cheaply and simply as possible. And it’s great practice to think of it - it should be absolutely thrown away when you’ve tested it. And then if it works out, then we replicate it. And then we might clean up the worst of it and make it a little bit more prepared for the future. But still very, very cheap and effective ways of testing things out.

And once it’s been proven that it works multiple times, then we start to formulate it. And this is by far the most difficult step. It’s basically going from something that runs on intuition and “maybes” to proving it out with data and finding out how it works. And then once we get to the point where we have the recipe, then usually it lends itself to be fully or partially automated.

And then we invest in making it. And this is key to how we captured all that complexity was because once we figured out the correct sequence of things to do to launch on the app stores, then it became a single command that fired off 20 other commands that did all the right steps in the right order.

Roland Siebelink: Yeah. And it sounds like this feels a little bit like a funnel, right? That there are a lot of projects probably that never even make it to the formulate stage or even to the replicate stage because you’ve already filtered out what didn’t work. That’s awesome. You say it’s called for “four eights?” Just getting the terminology right, Brian.

Brian Meidell: The four ates, as in the formulate, replicate.

Roland Siebelink: Ah, four ates. I got it. Okay. As in the past tense of eating, right? Very good.

Chris Benjaminsen: And the replications, that is something we purposefully put in there because you might do something and be lucky. But luck is very hard to replicate. I find a good way of always winning the lottery, right? It’s no longer a lottery.

Roland Siebelink: This is rather than just running one experiment, you actually want to have some solid significance behind your statistical test. Yes, exactly. Okay.

Brian Meidell: In the beginning of the process was more like validate, arguate, and then automate.

Roland Siebelink: That’s exactly right. You got from three ates to four ates, and that seems a little bit more solid. Why is the formulate stage so difficult?

Brian Meidell: It’s basically where you go from being lucky to being deliberate. I think it’s very - often you’ll find people that are talented, that do things for whatever reason. And then it works. I think Chris is amazing at balancing progression and games. If you play a game from Chris, it starts out easy and then it gets more difficult in the right way. It’s a very subtle art to be able to do that. There’s no real recipe for it.

And if you were to have someone else build those games, they don’t feel balanced in the same way. You either make it too difficult too quickly or not difficult enough too slowly or whatever. It’s just something that requires a lot of intuition. And now we’ve never been able to formulate that specific thing. But the basic idea is when you’re forced to make a recipe for someone else to execute on how you did something, then you suddenly get away from having individual people getting lucky or having really great intuition about something to getting to a point where you actually have a chance of scaling it.

Chris Benjaminsen: I think all companies actually intuitively decide what they want to work on next. We just turned that into a process, right? Most people basically start in the formulate step. You sit down and you write a product spec, right? And then you say, how do we measure this? What are our KPIs for this new product that we’re rolling out? And what does success look like? And whatever.

And very often we found that - this is very true in games - that’s a very costly exercise in comparison of just trying it. If you have a great idea, just try it. And then you can be lucky and it works. If that’s the case, if you’re seeing an improvement in the product as a function of this intuitive idea, go do it again. And then if you’ve been lucky twice, then it’s worth actually spending the effort. And that means you can do - you can possibly run maybe even two orders of magnitude for more experiments than a team that runs in a more standard way of operating.

Roland Siebelink: In terms of the hiring, Brian, what’s the positions that you’re looking for most at the moment and what kind of people fit FRVR the best?

Brian Meidell: Well, we have a very strong core of techies. But we like people who are strong technically and who have market sensibilities. We’ve been hiring pretty aggressively in the engineering department because traditionally we’ve had a very tiny engineering department who has managed to do a lot in a short amount of time. But now we’re getting to the point where we want to be able to do many more things in parallel. And we validated the unit economics of what we do and the basic system. Now we’re ramping up to be able to capture more opportunities in parallel.

Roland Siebelink: Value-wise, fit-wise, personality-wise, what’s the perfect FRVR employee.

Brian Meidell: I think intellectual honesty, a really strong technical side or whatever they’re supposed to be handling, they have to be fantastic at. I feel like we’ve gotten some really good people there.

But the curiosity about breaking the mold and trying new things and finding a way to solve problems that haven’t necessarily been found before rather than be unhappy about lack of documentation or beta technologies or a channel that has just started where everything doesn’t work properly.

I think that there are people who attack that as a challenge and who absolutely love succeeding at that. And then there are people who sit down and wait for the next release where the bugs are fixed. And our employees are squarely in the former camp.

Roland Siebelink: Okay, that’s a great description. Thank you for that. And just, in general, I’m very excited to see what’s coming down the pike for FRVR, especially your new round coming up as well. And again, if there’s investors listening who want an intro, please reach out. Brian Meidell, CEO, and Chris Benjaminsen, the founder of FRVR, thank you so much for joining this week. This was a true pleasure.

Brian Meidell: Thank you. This is a lot of fun. Take care.

Roland Siebelink: Thank you.


Roland Siebelink talks all things tech startup and bring you interviews with tech cofounders across the world.

Our coach has a very effective style of facilitation and introspection injecting provocative points into the contexts that leade to valuable extensions to the discussion.

Alan Young, InRule, Chicago, IL