In January of 2017, I quit a web application developer job to become a full time independent game developer. This followed nearly five years of development on my first game, while I balanced a full-time job and numerous tasks associated with building an independent game company. This is my year-long story of hopes and failures.

I am a 35-year old Canadian with a college degree in computer sciences and 13 years of professional experience. I worked with web applications for most of my life, but had a three year stint as a flash game programmer in the mid-2000s. Until starting development of my first indie game, I had never worked as a project manager and never developed a game on my own.

It's likely you've read bleak tales of indies launching their first games and failing in the market, and unfortunately, my story is equally grim. While indie darlings like Stardew Valley and Undertale manage to snowball into selling millions of copies, the average game is lucky to sell 1000 copies in their first month.

In 2017, I released two games on Steam. The first was Arelite Core, a retro-style JRPG released on February 8^th after four and a half years of development. Over 85 000 USD in production costs went to contracting artists and a music composer. Released only for Windows, it has thus far made less than 5000$ across all distribution platforms.

My second game was Astral Traveler, a runner racing game produced over seven months and costing around 15 000$ to develop between contracting a 3d artist, living costs and revenue sharing with another developer. Launched on September 13^th, it made well under 1000$ in game sales since, with approximately 12% of sales coming from Mac computers and 8% from Linux.

I also received approximately 2000$ from donations during my game development streams on Twitch, along with from a handful of generous patrons on Patreon. This brings my income total for 2017 to around 7000$, against yearly expenses of around 13 000$.

These sales numbers are nothing short of disastrous. Thankfully, I had planned for the worst before quitting my job, saving some money to ensure I could sustain myself for the year regardless of sales. But I won't deny that covering only 6% of the development costs for my two commercially released games is far from what I had expected.

To afford these low wages, I live a fairly spartan lifestyle. I have no spouse, no children, no car. I don't smoke, nor drink alcohol, and I rent a cheap two-room apartment. I cut out all luxuries, such as going to the theater and coffee shops. I keep my groceries to cheaper brand and, of course, video game purchases are off the table. And because my income is directly related to releasing games, I work between 70 and 80 hours per week, every week.

So, I want to take an honest look at one of the most difficult years of my life, explain my mistakes and faulty assumptions, and do what I can to help other hopeful developers avoid my pitfalls. I also hope to give gamers a better understanding of the process behind the creation of their favorite games, and the turmoil and uncertainty that follows.

But before diving into year one proper, I'll provide an overview of the development of my first game. Don't worry: This one is equally rife with costly mistakes, delays and life lessons.

Perception trumps reality

Arelite Core was unveiled at a bad time, following a large number of cheap titles produced using RPG Maker. Despite my best attempts to showcase its unique attributes, it was always perceived to be yet another throwaway 2D JRPG, and even erroneously has the "RPG Maker" tag on Steam. I believe this notion lost me a lot of potential sales.

In reality, I crafted a custom engine specifically for it, which added six to twelve months of development, while in the end providing very little perceived value to players giving it a quick glance. It used Microsoft's XNA Framework which would have made it compatible with the X-Box 360, but delays in production and Microsoft's discontinuation of the framework made those plans void.

Yet my engine was capable of a lot more than what RPG Maker had to offer, starting with a heavily customized battle system, taking inspiration from fighting games to inject a greater sense of strategy. I also added a crafting system, dynamic on-screen enemy encounters, a unique looking UI, and even diagonal movements on staircases to add a sense of depth to the world.

As with many first-time game developers, my goal was to create a grand and diverse experience with a multitude of different locations to explore and a large variety of characters and monsters to discover. I wanted to bring that nostalgic feeling I had playing the SNES Final Fantasy titles, while updating the genre for modern audiences. Instead of grinding through endless battles to move forward, the game would rely on its story for longevity, underlined with a cinematic presentation and unique, memorable soundtrack.

This ambitious scope gets to why the game cost 85 000$ to develop. It also underlines one of the most difficult aspects of creating a game like this: taking on a producer's role, as I hired and managed a team of artists to produce the numerous required art assets.

At the height of the game's development, I was communicating with at least five different groups. My role was to provide lengthy descriptions and reference material of the work for them to create, which they would translate to static backgrounds and frames of animations for characters, monsters and visual effects. I would then proof-check each asset as they were produced and test them in-game to ensure they matched specifications.

Over the course of four years, I contracted more than twenty artists to work on various visual elements, a process which involved putting out advertisements, looking over dozens of applications each time, interviewing candidates, writing contracts and of course, regular management to ensure work was done properly. This was a colossal task for one person developing games part-time alongside a full-time job and who had his own content to produce as well.

It's easy to underestimate the work required to manage a team, and it made me realize how difficult it is to find reliable workers, even if they’re being paid. Several artists I contracted were part-time hobbyists, so updates would often be sporadic. Many of them underestimated their production time frame, which led to them quitting the project, meaning they had to be replaced, starting the whole process again.

I originally intended to launch the game in 2015, but only received the final art assets in mid-2016. I underestimated the time to produce my own work as well, between programming the engine, creating the levels and designing and integrating the story scenes, which could take up to a week per scene. As time went on, Arelite Core weighed heavier on my shoulders as I left friends by the wayside to focus on bringing the game to its proper conclusion.

Then came PAX East 2016, one of the biggest reality checks for the whole project. I was offered a space in the minibooth section of the Indie Megabooth area. This gave me the opportunity to meet many industry veterans during an organized meet-and-greet, without a doubt the most important event of that weekend.

However, the lukewarm reception from attendees and press made it obvious that Arelite Core could be dead in the water. People didn’t hate the game, and it's difficult to appreciate a retro RPG in a convention environment, but I wasn't receiving the big burst of enthusiasm that the project needed to thrive on the gaming scene. The trip cost over 4000$, meaning that sales of the game barely covered it.

Following the convention, I started to stream Arelite Core's development on Twitch, 12 hours a week. Not wanting to spoil the story, I stuck to early sections and nondescript areas of the game. Growth for my channel was (and remains) slow, since little promotion is made for the creative sections of Twitch. Still, I managed to build a small, dedicated community around Dragon Slumber, which became invaluable as I transitioned to a full time career. For Arelite Core itself however, it was too little too late.

As I was anticipating the failure of my first project, I was still hopeful of my prospects for my second game. Now that I had more experience, I expected to be able to release a sustainable game as my next project. Confronted with my full-time job providing few advancement prospects and having never received a promotion in twelve years of career, I decided to gamble my future on becoming a full-time independent developer.

I planned several months ahead to ensure I could sustain myself for the year, mostly by banking as much money as I could and cutting luxuries from my life. I also began pre-production on my next game, which I developed with Bora Genel, a Turkish indie developer. He started creating the foundations in Unity while I finished development on Arelite Core, which was set to launch on February 8^th, a date selected due to the relative shortage of releases that week.

Christmas came, followed by the New Year and in January 2017, I quit my job. This is where the story truly begins.

Becoming an Astral Traveler

The transition to a full-time game developer was immediate as I ramped up my weekly working hours on Arelite Core from 40 to 80, a crucial change to finalize development and preparations in the last three weeks before launching. Up against a deluge of other games fighting for media attention, every aspect of my presentation had to be top-notch and I took care of lingering bugs, wrote press releases sent out press builds.

February 8^th rolled around and I launched Arelite Core live on stream. It was the culmination of four and a half years of work, yet I never had time to a appreciate the event. My next project was already under way, and I spent any extra time quashing the few bugs being reported.

Players who tried Arelite Core loved it, but as expected, the sales numbers were catastrophic. Now nearly a year later, I've barely sold enough to cover the costs of going to PAX East to promote it. Thankfully, I had anticipated those numbers, so they didn't affect my morale too much, instead focusing on my sophomore game.

Astral Traveler was a direct response to the issues which had plagued Arelite Core's development. To keep development costs to a minimum, I planned for a short 3 to 4 month development cycle and designed the game to require few original assets, instead focusing on gameplay. By keeping costs low, the game should be able to sustain itself once bundle sales are accounted for. Dividing tasks with my development partner would help keep things running smoothly.

Player competition would be placed front and center to keep player engagement high, through traditional leaderboards and even displaying friend scores during a race. And to spur on that competition, And finally, Astral Traveler would be released at a bargain price of 5$ to entice gamers outside of promotions.

I needed to do a better job with outreach this time around, as the secrecy for Arelite Core had made it more difficult to conjure up interest for the project. All of Astral Traveler's development would be shared early and often through Twitter and my development streams. I joined the CFN community, which helped me meet more industry professionals. In fact, I spent a significant amount of time improving communication in general.

I also made sure to always make time for the small but dedicated community which was building around Dragon Slumber, through Discord, Twitch, and direct contact on Twitter. By connecting directly with players and other developers alike, work on my own project improved and it even opened up new opportunities.

Access to that community helped me get faster feedback, both through comments made while streaming and direct testers of my games. I was also lucky enough to get some Twitch subscriptions and donations, along with some Patreon subscribers, which accounted for over 25% of my yearly income.

By March, Steam announced they were sunsetting Greenlight with no indication of the shape or costs its replacement would take. Not wanting to take any risks, we prepared a quick trailer and screenshots using early assets. The result was a game considerably different from what we intended to launch, both from an artistic and gameplay perspective.

It's natural for a game to evolve during its development, but many changes also grew out of necessity. We wanted to give it an abstract, flashy look, like a more organic version of Tron. Finding the right artist at an affordable price was more difficult than anticipated however, and after months of research, we decided Bora would establish the visual style and prepare various assets for the five worlds in the game.

We hired a 3D modeler to provide a distinct visual identity to the astral traveling ship itself, along with two enemies. Although the result was stunning, the production took a lot longer than expected and left us with higher costs than anticipated. As of this article, the game’s profit has yet to cover that production cost.

By the summer, we were several months behind schedule. The game's 47 levels were taking longer than expected to produce, while Bora balanced the unanticipated additional task of being the game's lead artist. We found that our visions for the game's design had diverged. Bora was more interested in a rapid, free-flowing experience, whereas I was driven by a challenging approach with quick player reactions. With the project already over budget, we quickly settled to go with my version.

We set a launch date for September 13^th 2017, seven months after starting production. Numerous proposed features were discarded, such as a VR version. I was hopeful for a mobile version, but due to our PC-first approach with graphics and controls, that would involve a considerable amount of extra development. We decided to wait and launch a second phase with those features if the game was successful. Likewise, we would consider console releases should we find traction on the PC.

Like Arelite Core before it, I launched Astral Traveler live on stream in front of people who had watched its entire development. Everything was in place for a more successful game, our outreach was better, the game was attention-grabbing, and it offered a great challenge in an environment where hard games like Cuphead were getting a lot of praise and conversation.

And then, it sold far, far worse.

Learning from my mistakes

Astral Traveler's failure hurt me more than Arelite Core. I had known for nearly a year that Arelite Core wasn't gaining the proper momentum, and although I didn't expect Astral Traveler to be a viral phenomenon, it still felt like it was tracking a lot better.

Yet despite my best efforts, it was released to very little fanfare, which I attribute to several factors. Although Astral Traveler incorporated a few genres, it couldn't stand out on the surface among other racers on the platform. The scope was too small for the average Steam gamer, and following the deluge of games unleashed by Steam Direct, cheap game bundles and free games released on a regular basis, it never stood a chance.

I spent the following month streaming development on Arelite Core: Lleana's Journey, a free expansion which used some assets cut from the main game to add an extra story segment and an additional difficulty mode. The update received a few sales and an additional review, but couldn't cover the cost of development.

Both of my games suffered from never getting a big break from influencers. I reached out to several big streamers and youtubers but never received a response. Likewise, big audience gaming websites ignored my numerous requests, which made it difficult to reach the gaming audience who might be interested in these games.

I decided that my next game needed avoid pre-established genres and expected themes. And because I would be working on this alone, I needed a visually minimalistic design which I could execute myself, focusing on simple presentation and novel, even polarizing gameplay.

A month after launching my second game, I started streaming development for Tech Support: Error Unknown. Presented through a Windows-like UI, it places the player in the role of a technical support specialist, chatting with customers, trying to resolve their phone issues while navigating corporate blackmail and hacktivist groups.

The gameplay involves the player interacting with procedurally generated customers through chat windows, using a series of macros to help conversation move faster. They earn more replies while progressing through the story and can even hack the system through the terminal command prompt to open new opportunities.

Tech Support offers a more open-ended experience than my previous games, with players free to engage in the conversations the way they choose, pick their allies accordingly and explore their computer environment to discover various story threads. It'll provide replayability and encourage several runs through hidden secrets, diverging paths and multiple endings.

It also has a more procedural structure rather than hard-coded levels, which makes balancing the game faster to test and easier to implement. I'll be using more tools to gauge player interest and issues, with the upcoming closed alpha including Unity analytics as well as a handy bug reporting feature using Trello's API.

2018 and onwards

2017 was by far the most difficult year of my professional career. I made 20% of my previous salary, launched two games that ultimately bombed, and transitioned from 30 to 80 hours a week of game development without gaining much more personal time.

While it's nice to avoid the freezing Quebec weather and having a looser sleeping schedule, I still work over 12 hours seven days a week, crunching on what I hope will become the next big thing. I have seen many reports repeating the importance of not overstretching and avoiding a burnout, but that comfort isn't a luxury I can afford yet. Owning a business isn't like a regular salary job, you don't get paid unless you make sales.

For the veterans out there, you don't need me to tell you how difficult this industry is. We're driven by passion, united as developers, but also competing against each other for our audience's attention, gamers and influencers alike. Those who keep their heads down in their game's development are due for a rude awakening. Great games are released all the time, and often lost in the tidal wave of other games.

Still, I feel privileged by my situation as a starting indie developer. I've had support from family and friends, including my stream viewers and people who generously donated to my Patreon. These sources have helped me afford to stay the course and have become my primary source of social interaction. Nevertheless, I must admit I feel like I let people down by having such a bad first year, and I can just push myself further in 2018 in response.

Failure is a reality for many starting projects, and an opportunity to learn and grow. In lieu of sales, I met many industry professionals who are helping me plan for my upcoming game, and releasing Arelite Core and Astral Traveler was instrumental to that. I'm also jumping on to new opportunities, like appearing on the Brightlocker platform for additional exposure.

So, get ready. In 2018, my voice will be heard.

And if you want to know more about Tech Support, feel free to join my newsletter or follow me on social media:

Dragon Slumber: http://dragonslumber.com
Facebook: https://facebook.com/dragonslumbergames
Twitter: https://twitter.com/dragon_slumber
Twitch: https://www.twitch.tv/dragonslumber/