Why I Left My Awesome Job to Build Keewano: My Beautiful Struggle

Many people believe that working at a big company is the safe way to go. For more than a decade, I had the opportunity to enrich millions of people’s lives at Plarium, working on some of the industry’s biggest titles.

I got to work with many incredibly talented people and learned so much along the way. But I grew to realize that, despite this great job and the security it brought, there were serious struggles that came with it.

In hindsight, these struggles could have been avoided (or at least alleviated) if I had the right tools at my disposal. And that was the main reason I sacrificed that incredible position to start something huge with Keewano.

The Biggest Challenge I Faced in Game Development

Game developers, if you’re reading this, you don’t need me to convince you how grueling our profession can be.

Of course, one of the most punishing aspects of the job is developing something you put blood, sweat, and tears into. And then, people just turn around and say, “Well, that sucked.” And you don’t understand why they felt that way.

Or maybe they did enjoy playing it! And yet still, you have no idea what it was exactly that they liked about it.

Up until now, the only solution we have found is to collect a ton of data and try to answer the following:

• Why are they paying here?
• Why are they churning there?
• Why exactly did they take this path and get stuck there? Maybe they’re missing something.

This applies to AAA games and free-to-play games alike.

This kind of uncertainty, just guessing your way through understanding your players’ behavior? It’s just a recipe for disaster, at least in terms of the decisions you might make to optimize your game.

And you’re wondering why one in 10 developers were laid off in 2024? Why game studios are closing – left, right, and center. Trust me, I’ve been there. I’ve had the terrible task of having to fire many wonderful employees for reasons outside of our control.

It eventually got to a point where it hit me and my co-worker, Dima Karger. We had a eureka moment where we realized how we could solve this critical problem. Let’s dive a little deeper.

The Limitations of Traditional Analytics

A huge reason for the suffering I endured in previous positions was the frustrations I had with old and current data analytics tools.

Look at it this way: To understand what players are doing, we have to collect everything they’re doing. This includes every single tap and every single interaction with the app.

Unfortunately, though, we have this limitation of existing databases. Because everyone can tell you we can store everything. However, retrieving this data is extremely expensive. This especially applies if you have a large audience of players. In other words, lots of traffic.

In my previous experience at Plarium (and most other studios and companies I’ve worked at, actually), we’ve tried creating a lot of predefined dashboards and also a lot of ad hoc reports by the teams of talented data scientists and analysts.

This would include all of the metrics that are most important to us. We’d then analyze them but ultimately end up just making a bunch of assumptions.

A lot of predefined dashboards and also a lot of ad hoc reports by the teams of talented data scientists and analysts

In short, if I see X and then I see Y, then there is probably some correlation between them. At that point, we’d:

• Build a hypothesis.
• Verify it
• Try to provide it
• Try to deny it.

So you’re iterating a lot of stuff, and that’s cool. But it’s full of suffering. It’s extremely time-consuming. This was the “Aha” moment for us.

The Turning Point

This was the defining moment when Dima and I started to seriously consider:

• “How can we speed up this process?”
• “What if, instead of just making an assumption, we can see the entire picture at the same time in some comprehensible way?”
• “What if I could see all of my millions of players and understand every single action in some kind of simple visualization?”

We saw great potential to create something better for game developers to eliminate that guesswork we agonized over for years. To find actionable insights – fast.

So we ended up just saying, “We need to develop this product.” And that’s exactly what we did. We called time on our incredible tenure at Plarium and started working hard to turn this idea into a reality – outside of the company, from scratch.

Building Our Game-Changing Database

To make this vision come to life, we had to develop a totally new database that was faster and more efficient than anything else currently available.

For comparison, the closest database in speed that we found on the market was Clickhouse. While not specifically for game developers, this database can process 256 million events in about 5 minutes.

And yet, the database we ended up developing managed to process the same amount of events in under a second – that’s 600 times faster!

This breakneck speed is a huge enabler. It allows us to execute extremely fast iterations and create our platform, Keewano.

The Future of Game Analytics

In future articles, I can’t wait to share with you exactly what our revolutionary database means for Keewano’s capabilities.

In the meantime, we are excited about its potential to transform how game developers understand their players and make better, quicker decisions.

I’ve been through a lot over the last few decades. There were many times I wished I had the platform we’re building here at Keewano.

Now, I just can’t wait to share this gift with game developers around the world to ensure they don’t have to face the problems that I endured for so long.