Marketing your first indie game — What we learned from releasing the same game twice.
For those of you who don’t know me, I’m one of the developers for Scarlet Hollow, an episodic horror visual novel that I co-create with my spouse Abby Howard. We started work on the game in March of 2020, with the goal of releasing our first episode as a free proof-of-concept in the fall of that year and using that to catapult into a Kickstarter to fund the rest of the game.
On paper, the launch went pretty well for an indie game — our Kickstarter overfunded by quite a bit, and we hit all of our deadlines. But there were also a lot of missteps with our strategy that wound up positioning us poorly for long-term success, so we wound up re-launching our Steam page earlier this year. It was a big risk, but one that wound up paying off.
This article is a timeline and analysis of our thoughts and strategies — our ideas, both good and bad — from the moment we started work on our game until now. Hopefully it can be of help to other indie devs out there, and if not, hopefully it’s an interesting read! And if you’re not interested in our beat-by-beat thoughts and are just here for the takeaways, feel free to skip ahead to the analysis.
Before getting started, I’d like to give a bit of context for our credentials and general metrics:
- We earned ~$95,000 in pre-orders between Kickstarter and BackerKit
- As of the time of this writing, over 30,000 people have our free-to-play first episode in their Steam library, and 13,000 of those folks have played Episode 1, with the median player finishing that Episode at least once, and the mean player finishing twice.
- The full title is currently in Early Access, and Episode 2 of 7 released about a month and a half ago. To date, we’ve moved over 5,000 units of the paid game across all of our distribution platforms.
I’d also like to acknowledge some of the unique advantages we’ve had on our side:
- Abby has been a full time cartoonist and graphic novelist for the better part of a decade, and has a large online following with a track record of financially supporting her work. This was a boon to our marketing and made our project a lot less risky.
- Abby and my backgrounds make it so we only need the two of us working full-time on our game, which means that we need less revenue to sustain ourselves. Both of us are writers, Abby is an incredibly prolific artist, and I have a background in business and marketing (and was able to quickly pick up the coding we needed to actually make the game — thanks, high school computer science!)
Planning and (secret) development
We decided a few things about Scarlet Hollow’s release strategy pretty early on. By the time we’d finished outlining the broad strokes of our story, it was pretty clear that the narrative already fell neatly into an episodic format. Scarlet Hollow takes place over seven days, with each day containing a partially self-contained story that fits into the overarching narrative and ends with a cliffhanger.
Going with an episodic release schedule had some clear upfront benefits. By necessity, it broke our work into easily digestible milestones that we had to hit, which in turn kept us guarded against feature creep. It also promised the potential of a steady stream of income over the course of development, which hypothetically meant we’d be less reliant on getting every penny we needed to finish the game from our Kickstarter or working with a publisher.
It also meant we’d have seven different opportunities to essentially launch or relaunch our game, and each of those launches could benefit from the mistakes made and the lessons learned from the launches that came before it.
As someone who spent a lot of time on the Life is Strange and Mass Effect message boards, releasing episodically was personally exciting, since Scarlet Hollow’s slow burn mystery felt like a perfect opportunity to create a similar community. It’s going pretty well so far — feel free to check out our Discord!
Because of Abby’s online following, we decided early on that we wanted to keep the development of Episode 1 a secret. We thought that announcing too early would be a bad idea — that we’d lose hype and momentum quickly, and that some people might have been dismissive of a graphic novelist announcing a pivot to games without having the goods on hand to back it up. It went against conventional wisdom, but it felt like the right move at the time. In retrospect, I think this was probably our first mistake.
So instead of announcing and leaning on drip marketing, our plan was to tee up Episode 1, and release it for free the same day the game was properly announced, with some vague lead up in the weeks before that to build up a sense of mystery and intrigue. We’d then release the remaining seven episodes as paid DLC, either through a season pass or sold individually, following the model of games like Life is Strange or anything from Telltale.
That’s not to say we did no marketing over the course of Episode 1’s development. For most of the six month development period, we actually live-streamed spoiler-free work on our twitch channel. Most of these streams were the two of us chatting with viewers while Abby hand-drew backgrounds for “a secret project” (Scarlet Hollow). We also did some game streams to quietly expand her brand in that direction.
One of my biggest marketing mistakes during this period was leaning too heavily on Abby’s online presence and not doing enough research into how people usually market games.
This started to make itself clear when I started on our Steam page and hit a frustrating wall: you needed your Steam page to be live for at least two weeks before you could release a build. I tried to explain our launch strategy to a Steam rep to see if there was a work around, but they insisted we follow their rules. This resulted in a lackluster page with minimal information sitting around collecting dust for a few weeks before our launch, which, as I realized in hindsight, penalized us in the eye’s of Steam’s algorithm. During that period, our bare-bones page averaged 3 wishlists per day.
Thankfully, Episode 1’s launch was successfully buoyed by a few things going in our favor:
- Our teaser trailer, which we put out without any context a few days before release, got about 50,000 views on Twitter.
- Our launch tweet performed pretty well too.
- We’d been doing a lot of behind the scenes reach outs in August and September to streamers and game journalists who were incredibly responsive to our pitch.
In our first week, Episode 1 of Scarlet Hollow received ~2,200 downloads and another 1,800 wishlists, as well as nearly universally positive user reviews, and that fed directly into a successful Kickstarter. It’s impossible, of course, to benchmark this against strategies we might have tried, but some things clearly worked, and, as is the case with most things, some other things probably could have been done better.
One of the challenges we would have faced had we revealed Scarlet Hollow earlier would have been juggling making our marketing materials with developing the game — our trailers, for instance, pretty much weren’t ready until it was already time to release. Maybe, even though we were going against conventional wisdom, one of our better decisions was coming up with a plan and sticking with it.
Despite that early success, it wasn’t long before we started to run into challenges with our long-term strategy— especially when when we started putting serious thought into how we’d deliver the rest of the game. Most of these problems were related to making our original page for Scarlet Hollow free.
- First and foremost was the realization that a free page with paid DLC would, by default, have reduced visibility in Steam sales, which felt like we were shooting ourselves in the foot.
- We thought about selling DLC both piecemeal and as a season pass, but the way we built the game would’ve been incompatible with selling piecemeal episodes on Steam.
- Changing our Steam page from free to paid would wipe all of our reviews.
By the time we were confronted by these problems, we’d already accumulated over 200 reviews with a 98% positivity score, and we were reluctant to do anything that might undo that progress. So we opted to set up a single DLC page for the game branded as a “season pass” which would update with new episodes as they released.
The new page performed poorly, probably in part due to confusion over just what a season pass meant here, and what would be included. It peaked early on at 17 daily wishlists a couple days after it was announced, and averaged out at around 4. And there was one day outlier.
We were stuck between a rock and a hard place — our free Steam page was an incredibly valuable marketing tool, but it felt like we were stuck on a course of action that would inevitably sink us, and it was a course of action that Steam advised us to stick with.
Around this time, though, we had the remarkable fortune of seeing other games in our free-to-play-page-for-their-first-chapter bucket decide to take a risk — Tiny Bunny was another hand-drawn horror visual novel that launched its first chapter for free, and they wound up launching a new Early Access page for their full game. And despite the remarkable success of their prologue, the new page performed incredibly well, quickly catching up to, and eventually surpassing their original page.
Relaunching Scarlet Hollow as an Early Access game was an idea that had been lingering at the back of our minds — it was the decision we would have made had we been able to go back in time and start over with the knowledge we’d learned from our first launch. It would get around the visibility issues we were worrying about and clear up most of the confusion around how we were actually selling our game. Seeing another developer take that risk for themselves and survive was the final motivation we needed to do so ourselves. So we queued up a big marketing push to coincide with revealing the new page, including a set of wishlist milestones to reward our community for supporting us.
Reception to the new page was incredible. We received over 1,000 wishlists on day 1, which diminished to a daily average of ~60 after a few weeks, still an order of magnitude better than what we were seeing with our season pass DLC. This led to our new page showing up towards the top of a lot of tag-specific upcoming games sections, which fed back into our visibility.
A lot of that game from how we teed up support on Day 1 — we gave all of our followers and backers, across all of our platforms, a heads up about what we were doing, and we were able to make sure they showed up in numbers. Our new page was also supported by months of work we’d done optimizing our free-to-play Steam page’s marketing, from its capsule images, to its copy, to its tags.
In less than a day we had more wishlists on our new page than we had for our DLC. By launch day, which was a mere couple of months after the page went live, we had nearly as many wishlists as we had for Episode 1, which had been up for 7 or 8 months. At the time of this writing, a month and change later, we have 3x as many wishlists as we had at launch, and 1 in 3 people who’ve played Episode 1 have converted into paying customers.
Takeaways and lessons learned
It’s quite possible that our Early Access page wouldn’t have had nearly the same success if we hadn’t launched Episode 1 as a free-to-play page as part of an accidental dry-run. That being said, here are some of the biggest lessons we’ve learned:
Don’t over-commit to a bad situation. Regardless of how much Episode 1’s release strategy might have contributed to the success of our relaunch, I think it’s safe to say that continuing to stick to our guns instead of pivoting would have hurt.
Look at what other people are doing. Keeping an eye on what other devs with similar games were doing and how it played out for them was what made us finally commit to changing our course. In particular, seeing Tiny Bunny pull off a successful transition to an Early Access page meant we got to see some validation for our go-to-market plan before committing to it.
Know when not to listen to other people. Steam advised us to stick to our original DLC strategy. I think it goes without saying that based on our visibility numbers, this would have resulted in a weaker launch.
Launching Episode 1 on its own page instead of as a demo probably helped. Steam charges $100 to list a game on its platform, but that extra $100, compared to a demo, wound up more than paying for itself in terms of access and marketing. To this day, over 30 people wind up playing our Episode 1 build daily, and there’s a clear funnel leading those players to wishlist or purchase our Early Access build.
As part of my market research work, I spend a lot of time looking at what does and doesn’t work for other titles by tracking concurrent player and follower counts on steam.db. Even though Steam puts considerable resources into NextFest, demos tend to get a small bump before falling off hard — and they don’t get their own separate visibility funnels. If you have the quality to back it up and manage to get good reviews, a separate Steam page with a free demo/preview of the full game can be an exceptional asset for evergreen marketing.
Having a separate page also meant we had a good opportunity to learn about our game, which is one of the most difficult and important pieces of marketing you can possibly do. As creators, we’re intimately familiar with every small detail of our work, but that familiarity makes it so much harder to see what we’ve made in broad strokes. And if you don’t know what you’ve made, it’s that much harder to sell it. Even now, this is an ongoing process for me. Scarlet Hollow is a visual novel, but it’s a visual novel that does very different things than most examples of the genre. It’s a horror game, but one where the scares are punctuation in a longer narrative about loss and regret that’s also funny.
While I’m still trying to figure out the best way to pitch our game to people, our soft launch gave us a lot of insight over what details work. When people know nothing about our game, are they more likely to check it out if there’s a picture of Stella on the cover? What if there’s a picture of Stella and her pug, Gretchen? How much should we lean on Abby’s name recognition?
Trying to answer these questions put us in a position where, by the time we launched our new page, we knew what worked on a micro level, which in turn meant that once we fed good data into Steam’s algorithm, page views and wishlists would continue to organically trickle in.
Final Thoughts
Over the course of writing this article, I realized that it didn’t have a core thesis. I thought it would mostly come down to how doing a surprise release with a free Steam page was a bad idea, but the more I had the chance to gather my thoughts, the more I realized that maybe it actually worked for us.
If there were a core thesis, it would probably be to regularly reassess your strategy— to constantly take a step back and ask yourself “why” you’re doing what you’re doing, and whether there’s something else you should be doing instead. To take risks and try new things instead of over-committing to something that might have worked once, but that wouldn’t continue to work.
And you might want to think twice about how you structure your demo.
Ultimately, every game is different, and there’s no magic formula for success. The most important thing you can do is to keep trying new things, keep seeing what works, and keep pushing yourself to figure out what you’re actually making.
If you want to see more posts like this, don’t forget to follow us on Twitter, and consider supporting our studio on Patreon!
