Streets of Rogue: An Indie Game Built on $18,000 Sold for Roughly $6.5M — the Six-Year Story

All dollar amounts are converted at roughly ¥150 to $1.

There’s an exit path where an indie game gets bought, IP and all, by a public company. The developer was a single person, Matt Dabrowski. Development costs, mostly outsourcing fees, came to about $18,000. What grew out of that was ‘Streets of Rogue,’ which sold roughly one million copies cumulatively, reached $8M in lifetime revenue, and was acquired by publisher tinyBuild in June 2021 for $6.5M (split roughly evenly between cash and stock). This is the story of a project one person started with a few thousand dollars turning into an asset worth hundreds of millions of yen — tracing where the numbers came from and where the turning points were.

Timeline and numbers

What was he building

‘Streets of Rogue’ is a roguelike action game where the player tackles a city using a wide variety of characters. Development was Dabrowski working alone throughout, and the $18,000 budget went mainly to outsourced music and art. In other words, he kept the core (design, programming, and overall structure) for himself, and only handed off the areas he wasn’t good at.

Two points where the trajectory shifted

There are two turning points in this story, and they’re different enough in nature that they need to be looked at separately.

The first is the free alpha release in October 2015. Rather than selling a finished product, he put out a free version to gather feedback and gradually build a player base. The Steam Early Access launch in March 2017 extended that same approach, settling into a pattern of developing alongside a core group of players. As a result, by the time of the full release in July 2019, there was already an audience that could supply early sales and favorable reviews. The 96% Steam rating with over 15,800 reviews is the product of nearly four years of runway, not something collected on launch day.

The second is tinyBuild’s IPO in March 2021. The acquisition approach itself had come in summer 2020, but tinyBuild going public at a $600M market cap suddenly accelerated the negotiation, which closed roughly 2.5 months later. The fact that the sale consideration included ‘newly issued stock’ directly reflects this causality. A buyer who had just gained access to the currency of public stock came to fold the IP of its highest-rated flagship title into the company. That’s the structure at play.

Something to flag: meeting tinyBuild at GDC in 2016 and cold-emailing multiple publishers became the starting point of a relationship with the eventual buyer, five years later. The turning point didn’t arrive out of nowhere. It was a connection he built himself five years earlier, finally ripening.

Breaking down why it grew

First, the structure invested time rather than capital. The $18,000 development cost was the result of keeping the outsourcing ratio extremely low. With no borrowed money, there was no repayment deadline and no investor demanding growth. That’s precisely how he could spend nearly four years from the 2015 alpha to the 2019 release. This ‘absence of a deadline’ bought the time needed to refine the work through Early Access.

Second, high ratings created the IP’s price. For tinyBuild, ‘Streets of Rogue’ was its highest-rated title. Publishers typically hold revenue-share rights but not the IP itself. Buying the IP means buying the right to make sequels and spin-offs at your own discretion, and a 96% rating is material that substantiates the future value of that right. It didn’t sell because revenue was large. It sold because it was supported enough that a sequel was worth making.

Third, the seller drew a clear line on what to give up. What was sold was the game’s IP. Dabrowski kept his own studio, DogHelm Studios, and set things up so he could continue developing other titles. Rather than selling everything, this was a design that carved out the single point where value was concentrated.

The numbers need careful reading

Dabrowski himself flagged a caveat about the $8M lifetime revenue figure: ‘Most of that amount went to the various Steam/Xbox/PlayStation/Nintendo stores, to tinyBuild, and to taxes. What DogHelm actually received was quite a bit smaller.’ The take-home after platform fees, the publisher’s cut, and taxes is a different thing entirely from the headline revenue figure. This is the part most easily misread when reading indie developer earnings.

The cost of the sale process itself wasn’t small either. He described the legal and contracting process as ‘long, expensive, and never particularly comfortable’, after bringing in a specialist familiar with game law and revising the contract multiple times, legal fees exceeded $70,000. On top of that, part of the consideration is tied to future revenue share and performance conditions linked to operating targets for the existing game and its sequels. The $6.5M figure was not, at the time of contract, entirely confirmed cash.

What can and can’t be replicated

The design philosophy is what’s most reproducible. Keep the core work for yourself, outsource only what you’re bad at (music, art), and keep development costs in the low millions of yen. Release a free version before the product is finished to build an audience, refine it through Early Access, then do the full release. When it’s time to sell, carve out just the IP and keep the corporate entity intact. All of this transfers directly to non-game individual products as well.

The conditions that don’t transfer are equally clear. The buyer’s IPO was an external event that accelerated negotiations. Steam Early Access is a market that lets you sell an unfinished product for money while you develop it, and that market has to exist. And the relationship built at an industry conference in 2016 paid off five years later. None of these can be summoned by effort alone. Beyond that, behind the one-million-unit figure lie countless titles that never got there, and a 96% approval rating isn’t a variable you can assume as a baseline for replication.

What this case demonstrates isn’t that ‘even a solo developer can build an asset worth hundreds of millions of yen.’ It’s the sequencing: only someone who builds a structure allowing time without injecting capital can afford to wait until the reputation solidifies.

Related reading

• Photo AI — a solo developer building an audience before the product
• ScrapingBee — the sale process behind a small-team product