From game development and bug bounties to a $40M/yr product

George Deglin was building games and struggling to keep players engaged. When he realized that he wasn't the only game developer struggling with this problem, he built the solution.

Now, OneSignal is bringing in over $40M ARR.

Here's George on how he did it. 👇

The first side project

I grew up in the Bay Area, which meant two things: There were always computers lying around, and it wasn’t weird to know someone’s dad had just started a company.

Both of my parents were software engineers, so tech was kind of the de facto family language. I started dabbling in programming as a kid, but more than anything, I was fascinated by the idea that it was possible to build something entirely new, given the right tools. 

I went on to major in computer science, but I didn’t stick around long. During the summer after my sophomore year, I started working on a side project with some friends. It started off casual, but like most side projects with too much potential, it quickly stopped behaving like one.

Getting acquired and starting up again

I ended up leaving school early to cofound and build that company, Uversity, where I was CTO. We built social networking products for universities, raised a few rounds of venture capital, and eventually got acquired by another company in the education space.

In the early days, though, it wasn’t exactly glamorous. As a handful of scrappy 20-year-olds trying to figure things out, we found the cheapest office space we could. It must have been the coldest, most poorly insulated warehouse in all of San Francisco. I’d literally show up in a ski jacket and fingerless gloves and sit there for 12 hours a day, hacking away like that was normal. None of us really knew what we were doing, but that was also kind of the magic of the whole thing. We had nothing else going on — it was all-in, feet to the fire (portable heater), learn as you go. 

After that, I met my current cofounder, Long. He’d previously started a social gaming company called Gaia Online and had this incredible intuition for consumer products. It was an instant connection and one that felt truly balanced. My background was more technical, his more creative — and together we built our own mobile gaming studio (Hiptic Games) right as the iPhone and App Store ecosystem were starting to take off.

That experience ended up pointing us toward a much bigger opportunity. One that would eventually become OneSignal.

Stumbling upon a bigger problem

OneSignal was born out of a very real problem we were facing firsthand. At the time, Long and I were running Hiptic Games, and like most game developers, we were constantly trying to figure out how to keep players engaged.

The biggest challenge wasn’t building the games themselves, but rather retaining users and building a relationship with them after they downloaded the app. Back then, there really weren’t any good tools for messaging your users. Push notifications were technically possible, but the infrastructure was brittle or non-existent. So we started building internal tools ourselves (push, email, SMS) just to keep our games alive.

The more we built, the more we realized this wasn’t just a niche problem for us. This was a fundamental need for every app developer. That’s when we decided to go all in and turn those internal tools into a real product.

That said, it wasn’t a classic “two experienced founders raise $5M on day one” kind of story. It was much more of a winding road.

Paying rent with bug bounties

At the time, I was not paying myself much, and one of the ways I was able to pay rent was by doing bug bounties — finding security vulnerabilities for platforms like Facebook, Dropbox, and Coinbase. At one point, I was the #2 highest-paid bug bounty researcher on Facebook’s leaderboard, which still feels a little surreal.

It ended up being great training. Doing bug bounties forces you to think about how complex systems behave. You might be staring at a system with 10,000 possible paths, but only one or two actually matter. That process helped me sharpen my intuition as an engineer and quickly zero in on the priorities that matter. It’s a mindset that served me extremely well when we started building OneSignal.

Since then, we’ve grown OneSignal into a platform used by over 2 million developers and marketers, surpassed $40 million in ARR, and now power messaging for one in every five apps released worldwide.

A not-so-minimal MVP

A lot of people assume successful businesses hit product-market fit early, but that was definitely not the case for us. In fact, one of the hardest things about building OneSignal was that our minimal viable product was far from minimal.

Most startups can get away with launching something simple. But in our case, we were building infrastructure for push, email, and SMS. The bar was much higher. 

When Long and I sat down to write out the list of what the product would need to do, it was comically long. On paper, it looked like something only a 50-person engineering team should take on. But we’d just come off years of game development, which is its own kind of brutal. We were used to writing absurd amounts of code under pressure, so while it wasn’t easy, we weren’t intimidated by the work. 

The advantage of scratching your own itch

There were also some advantages to being a tiny team. No big meetings, no approvals, no shipping roadmap. In our case, the only "stopping and thinking" was pausing for a minute to sketch something on a piece of paper, and then it was straight back to writing code. Before long, we had something real: a tool where you could create and send messages across multiple channels, build audience segments, analyze delivery, pull data from an API. From there, we layered on mobile SDKs to support iOS, Android, Unity, and more.

At the same time, we were talking to other developers — companies that were facing the exact same pain we’d felt. Those conversations gave us feedback, but more importantly, they gave us confidence. We weren’t just guessing. We’d lived this problem ourselves and we knew how painful it was. That gave us a real advantage: the product wasn’t built from the outside in, it was built from the inside out.

I think that kind of empathy is hard to fake. A lot of companies talk to customers. Fewer have actually been the customer. When you’ve sat in that exact seat, you know what actually matters and you’re more motivated to get it right. It’s hard to put into words, but I think that energy shows up in the product.

A reliable stack

On the frontend, we use React and TypeScript, with some Ruby on Rails as well. 

Where things get more interesting is on the backend. A lot of our core services are written in Rust, a systems programming language that’s known for being fast and extremely reliable. It does come with a steeper learning curve, but for us, the trade-off is worth it. When you’re delivering billions of messages across platforms, reliability is critical, and Rust gives us the performance guarantees we need at scale.

On the database side, we use a combination of PostgreSQL and ScyllaDB. One of the early technical decisions that made a big impact was how we approached storage. A huge part of our value proposition is giving customers real-time audience segmentation and event-based automation, which means constantly reading and writing large volumes of data. And if that process is slow, the product doesn’t feel responsive.

New tech opens doors

Around the time we were getting started, NVMe storage was becoming more affordable and widely available in cloud infrastructure. That gave us access to extremely fast disk I/O at a fraction of what it would’ve cost just a few years earlier. It might sound like a small thing, but it played a big role in helping us meet our performance and scalability goals, without having to pass massive infrastructure costs onto our customers.

In fact, that’s one of the reasons products like OneSignal didn’t really exist before. Not because people didn’t want them, but because the economics didn’t work. If we had tried to build this five years earlier, it might’ve been technically possible, but not financially viable. 

But that’s the magic of technology, right? It compounds. Every year, it gets just a little bit better, and that opens the door to new business models that weren’t viable before. We’re seeing the same thing now with AI: capabilities are improving, costs are coming down, and we’re starting to find really promising ways to incorporate it into our product, in ways that give customers a ton of value without breaking the bank.

Making the product an easy "yes"

In the early days, our first users came from our own network: friends, former colleagues, other game developers we’d met while building Hiptic Games. Some were people we’d worked with directly, others were folks we reached out to for feedback while we were still figuring out what OneSignal could be. A few of those early conversations turned into our first paying customers.

One of the big unlocks early on was a decision we made around SDK support. Most messaging platforms at the time only supported native Android and iOS SDKs. But a lot of mobile developers (ourselves included) use cross-platform frameworks like Unity, Flutter, or React Native. We knew from firsthand experience how painful it was to wire a native SDK into a non-native stack. So instead of launching with just native support, we made sure to build great SDKs for Unity, React Native, and other cross-platform environments. It required more work up front, but it paid off. It made OneSignal immediately usable for a much broader set of developers.

We also prioritized localization from day one. Having built games, we knew that distribution is inherently global. You launch a title and suddenly you have users in 50 countries. So when we launched OneSignal, we included support for collecting preferred languages and sending localized messages in the very first version. That made it easier for global apps to adopt us out of the box.

But the biggest growth driver, by far, has been our freemium model. Most of our competitors require you to talk to sales before you can even try the product. We took the opposite approach: anyone can sign up and start sending push, email, or in-app messages in minutes. No sales call, no credit card. It’s been huge for our growth. Every single day, hundreds of people sign up and explore the platform. Even if they don’t all convert right away, that kind of exposure creates word of mouth, and over time, that’s incredibly powerful.

Aligning incentives

Our free plan is surprisingly capable. You can send unlimited push notifications, get 10,000 free emails per month, set up Journeys, build audience segments, and access analytics. That’s been a really attractive way for developers, startups, and small businesses to get started. Some use it just to evaluate the product; others run their entire business on the free plan.

As companies grow or need more advanced capabilities, we offer a self-serve Growth plan as well as higher-tier plans with volume pricing, more sophisticated features, and dedicated support. But we’ve been deliberate about keeping pricing accessible, especially early on. We want the amount we charge to scale with the success of our customers.

For example, our push notification pricing is based on monthly active users. That means if your app grows, we grow with you. Our incentives are directly aligned with helping our customers succeed, and that’s shaped a lot of how we think about product and pricing.

Early on, most of our revenue came through self-service. But as we started working with larger companies, we built out a sales team and a customer success org to support more complex use cases. Today, we serve a wide spectrum of customers; from indie developers and early-stage startups all the way up to large enterprises, including major financial institutions, media companies, and government agencies.

That range is something we’re really proud of. Not many companies can say they support both a weekend side project and a Fortune 500 business on the same platform. But that’s always been our goal: to democratize access to world-class messaging tools. No matter your size, you should be able to build meaningful relationships with your users, and OneSignal makes that possible.

Advantages of living in the Bay Area

One big advantage for us has been being based in the Bay Area. It’s hard to overstate how much of a difference that’s made, not just in terms of access to capital or customers, but in the sheer density of talent and ambition. When you’re surrounded by people who are building, shipping, and pushing boundaries every day, it creates a kind of ambient pressure. In a good way — it keeps you extremely sharp.

There’s this deep-rooted passion for creating things, and people genuinely want to see others succeed. Whether it’s advice over coffee, intros to investors, or just a sense of camaraderie, that community has played a massive role in our journey.

Here's how to get started

The hardest part is building something people actually want. That sounds so simple, but it’s where most early efforts tend to fall apart. A good place to start is by looking at the problems you already have personally, especially the ones that feel repetitive.

If nothing comes to mind right away, try something new. Explore different spaces and go looking for those problems. The best ideas usually don’t come from that cliché flash of inspiration, but rather from simply paying attention.

Once you do pick a path, just know you’re going to be spending a long time on it. That kind of commitment isn’t always easy, especially in the early days when progress is slow and the work is heads-down. I remember plenty of weeks where it felt like all I did was write code, sleep, and repeat. 

The turning point is when you start to see even a tiny bit of traction. When people use what you built, and it actually solves something for them. For me, that energy still exists today. Even at this stage of the company, I get fired up thinking about new challenges, new hires, and new customer problems to solve. When you build something dynamic, the momentum kind of builds on itself.

My advice? Set real milestones. Pick a launch date. Ship something. Give yourself a target to work toward, even if it’s just a line in the sand. Whatever happens, you will end up learning something important from the outcome. The important thing is that you commit to the attempt.

What's next?

On the product side, our mission is to build something that has a truly transformational impact on businesses around the world. We want OneSignal to help millions of companies create better customer experiences, grow their revenue, and save time along the way. In doing so, we’re not just improving individual businesses, we’re helping expand the entire consumer economy. 

To realize that vision, we also have to continue scaling our business. That means growing revenue, hiring exceptional people, and continuing to invest in product development and customer success. If we want to build something that lasts, and really fulfills its potential, we need to be one of the most successful software companies in the world. That’s the bar we hold ourselves to.

One of the most exciting opportunities on the horizon is AI. It aligns incredibly well with our mission to democratize customer engagement. In the past, only large companies could afford to build out sophisticated messaging strategies, with teams of marketers, consultants, and tools to support them.

Now, AI gives smaller companies access to capabilities they’ve never had before (expert recommendations, automated experiments, error detection, campaign creation) all built into the product. Suddenly, a one-person team can operate like a fully staffed department. 

That’s where we’re headed, and I couldn’t be more excited about what’s next.

You can follow along on LinkedIn. And check out OneSignal.