Black Tabby Games — 2024 State of the Union

Hi everyone!

We’ve been pretty quiet about timelines and development milestones for a while, and the biggest reason for that is that we’ve had a lot going on that we couldn’t talk about. Until now! So I thought it would be a good idea to start doing an annual “State of the Union” (because State of the Studio sounded too dramatic) to share what we’re able to share, and to give an in-depth breakdown on the choices we’ve made as a studio and why, to the extent that we’re able to talk about them.

One of the (many) hard things about games marketing is maintaining the balancing act between sharing exciting news with your community and holding enough cards close to your chest to keep The News interested.

In this case, we’ve had a ton of exciting developments for Slay the Princess — The Pristine Cut, but we had to hold them back until now to make sure we’d have a slot in Summer Games Fest. In case you missed it, though, here’s the new trailer:

In short:

• The Pristine Cut got huge at some point (it makes the whole game ~35% bigger by most metrics.)
• We’re partnering with Serenity Forge to bring the game to consoles and do a physical release, including a Collector’s Edition.
• We’re localizing the game’s text in 12 languages!

At this point, our work on the game is pretty much done, and now it’s up to the porting team and localizers to do their thing. I’ll still be spending some time in the coming months as a resource for these folks, but for the most part, this has wholly freed us up to focus on Scarlet Hollow, which brings me to the meat of this update.

Why is Episode 5 of Scarlet Hollow Taking So Long?

All of you have been extremely patient with us about Episode 5’s development cycle. We deeply appreciate that patience, and would like to ask for just a bit more as we continue our work to finish the remaining three episodes. Fortunately, the success of Slay the Princess means we are able to work on Scarlet Hollow full-time, on our own schedule, until it’s done! And that is so, so exciting for us. It’s any creative’s dream to be given the kind of freedom that allows them to work on their pursuit, free from the worries of budget or external oversight, and we are so, so grateful that we are able to do this now, and hope you love what comes of it.

But we also wanted to shed some light on the thought process that went into the decision to briefly pause work on Scarlet Hollow and crank out a whole extra game, as we know it probably seems like an odd decision.

When we started work on Scarlet Hollow, we knew that the title had an extremely ambitious scope, but I don’t think we knew at the time just how ambitious that scope was. Most of this ambition comes down to two things:

1. The game has an enormous amount of art, and all of it is drawn by one person.
2. The game’s script is enormously reactive, and all of it is written and coded by two people.

We think that part of what makes Scarlet Hollow so special is the combination of delivering on its ambition while being made by a core team of two who fully share the creative vision.

Though this presents a pair of challenges. The reactivity pushes the game’s scaling in an exponential direction with each episode, which takes time for two people to implement, and the hand-crafted dialogue and art means we can’t scale the team without losing at least some of what makes it special.

To give a little sense of this scaling:

1. Episode 1 has ~100,000 words*, ~120 background images and 659 sprites. It took us 6 months from the start of development until release. *Note: due to Episode 1 being the first game I ever coded, a lot of segments are inefficient about handling repetitive text. I would estimate the “real” word-count to be ~70,000.
2. Episode 2 has ~100,000 words, ~90 background images and 951 sprites. It took us 9 months from the release of Episode 1.
3. Episode 3 has ~170,000 words, ~50 background images and ~2,000 sprites. It took us 9 months from the release of Episode 2.
4. Episode 4 has ~200,000 words, ~130 background images and ~1,500 sprites. It took us 9 months from the release of Episode 3, the last few of which were utterly brutal for the both of us.

We assumed, when we started development, that the experience we would gain with each new episode would offset the increased development time, and this was true until we finished Episode 3, at which point our ability to move faster plateaued. Which is nothing to be ashamed of! A 200,000 word visual novel with over 1,500 art assets in nine months is pretty good.

That said, as of the time of this writing, Episode Five is shaping up to be probably a little under double the length of Episode Four, meaning that, at our peak development pace, had we not worked on anything else, it probably only would have been ready about… now, actually (18 months for ~400,000 words, and Episode 4 released ~a year and a half ago.)

By the time Episode 5 is finished, Scarlet Hollow with have this many words in it, +- 20%

That said, there are a few things that would have made that reality impossible:

• Burnout. Abby and I have each been working 60–80 hours a week, every week since we founded Black Tabby Games, and Abby has been working that style of workweek for far longer than that. It’s been an unfortunate necessity to stay afloat in creative spaces, especially comics, but it’s also led to building burnout over the years. We were 28 when we started this, and are 32 now, and we can definitely feel those four years. Episode 4’s development was particularly brutal, and we needed a solid couple of months after its release before we were recovered enough to return to work. The pace has been unsustainable, and it’s taken a toll on our health, especially over the past couple of years. I started to go grey at 31, which Abby has been a big fan of, but concerning nonetheless!

They say you gotta go fast. But they never tell you the cost of Speed™

• Funding. Scarlet Hollow has performed very well for what it is: an early-access horror visual novel, which is a tough sell. And including its Kickstarter, ~80% of its lifetime revenue to date has been generated after Slay the Princess was announced, despite that period of time only accounting for about 40% of the time the game has been available for purchase.

Elaborating on the funding and revenue side of things, with rare exceptions, game sales follow a chart that looks something like this:

The first two years of daily downloads for Episode 1 of Scarlet Hollow. Yes, it’s a free-to-play game, but the concept remains, and is more baked in the attention-economy than sales.

That is, you move most of your copies in the first week of your launch, and then it dramatically falls off. The rough rule of thumb people use is this:

Launch Wishlists * 0.2 = Week 1 Sales**

First month sales = 2 * Week 1 Sales (includes week 1 sales, so essentially the next three weeks equals that first week.)

First year sales = 3 * Week 1 Sales (includes first month sales, so the rest of your first year will bring in roughly your first week of sales beyond the first month.)

** Note that this 0.2 is not a 20% conversion rate, and includes week 1 sales that happen without a prior wishlist.

There are exceptions to this rule, and there’s a wide range of variation within it. Some games convert at much higher than 20%, and they tend to have longer tails, meaning that sales fall off far more slowly. Others might still have a 20% ratio but with a longer tail. Other games still might convert at much lower percentages. I’ve seen a few games out there convert at 5% of their launch wishlists in the first week or less.

Here’s what the first year and change of sales for Scarlet Hollow looked like on Steam, not counting refunds (we’re lucky enough to have a far lower refund rate than the industry average for both of our games):

The spikes in this graph are discount periods. Some are associated with new episodes or content updates (the first big spike was when we released the Talk to Animals patch) and some are just seasonal sales.

Note that if you’re trying to do napkin math here, the majority of a game’s copies sell while discounted (Scarlet Hollow only ran 20% discounts before Episode 4) and we use regional pricing for our games, which means that the typical sale price is maybe 65% MSRP. After this, and after refunds and chargebacks, Steam then takes 30% of our sales as their platform fee, so all in all, for a game with a $20 USD sticker price, the average sale puts maybe about $10 into our company’s bank account.

Scarlet Hollow launched into Early Access with 4,500 wishlists, and we made 803 sales on our first week, bringing our wishlists to week 1 sales ratio to 18.9%, or a little under the average. We also had 1,053 people redeem keys from our Kickstarter campaign, which if you factor that into our ratio, brings it instead to 41.2%.

Redeemed keys from crowdfunding campaigns don’t count as sales for Steam’s algorithm, however, nor do reviews from those keys count towards Steam’s review averages, so while the game performed well above-average relative to its launch wishlists, the algorithm pegged its performance as below average, limiting our visibility somewhat. This is one of the main reasons we didn’t crowdfund Slay the Princess, and it’s one of main reasons we likely won’t do any future crowdfunding campaigns for games.

With Scarlet Hollow, we were fortunate enough that word of mouth kept the game alive. In its first month, we had 1,744 sales, slightly beating expectations, and in the first year, we had 6,503 sales, significantly beating expectations. That said, significantly beating expectations didn’t mean that our sales numbers weren’t dwindling, nor did it mean that we were earning enough revenue in general, since we launched with a low wishlist count. (Standard modeling would have projected about $21,000 in year one revenue after Steam’s cut and refunds, but before paying contractors — then split among two people, before tax.)

In the three months leading up to Slay the Princess’ reveal, after Steam’s cut and after paying contractors, Abby and I were each making about $5 an hour. While we had decent cash reserves leftover from our Kickstarter, that balance was going down each month, and while we were confident Scarlet Hollow’s eventual 1.0 release would be a big marketing beat—each new episode had a bigger release on Steam, and we know a lot of people are waiting for the full release to pick up the game—we also saw the writing on the wall: we wouldn’t be able to finish our work before running out of funding if something didn’t change. It didn’t help that during this time, we were working between 60–80 hours a week, and burnout meant that each week was physically getting harder than the last. It was a pace we could only keep up for so many years.

So we had a few options:

1. Market the game better. Scarlet Hollow has a weird marketing profile. It’s exceptionally hard to get people into the door for a number of pretty obvious reasons — it’s a visual novel (I love our medium, but they’re a harder sell for general audiences); it’s in Early Access, so it’s an unfinished game; the pitch is long, and snappier versions of it don’t convert. On the flip side, however, once people start playing, they love the game. We have an unbelievably low refund rate (1.8% vs industry standard of 10–20%), and have had a ton of word of mouth spread, which has kept us going through years of shoestring development (thank you all so much!! We love our community!) But we thought, maybe if we just marketed the game better, we could get more people through the door, and things would pick up from there. (This was our first choice, but none of the things we did moved the needle enough.)
2. Sign with a publisher and get external funding. This is what most people do, and we explored our options, but weren’t happy with any of the terms we discussed with people — publishers, in general, wanted too much revenue (usually 50% post-recoup), and would have tightened our schedule and influenced creative direction enough that the end-version of Scarlet Hollow would have, in our opinion, been much worse. Beyond this, negotiations with publishers were intensely time consuming, often eating up weeks of work. Given the extent to which publishers have shut down and shuttered projects over the past couple of years, I think we made the right choice here.
3. Do another crowdfunding campaign. If worse came to worst, we could have done another crowdfunding campaign, but of course that would be incredibly unwise considering the initial Scarlet Hollow campaign won’t be fully fulfilled until the game is done and artbooks go out. And even without that consideration, campaigns take an enormous amount of effort (it’s usually about a month and a half of full-time work from both of us to manage a campaign), and any new rewards would have had their own cost — bespoke rewards, like commissioned art, would take time Abby would otherwise spend working on the game, and merch has its own production costs. It would have been an all-around bad idea.
4. Do paid consultation on another project. This is a trap that a lot of young companies fall into. They do paid consulting work to keep the lights on while simultaneously working on their main project. The issues here are two-fold: you have to find this work, which takes time; you have to do this work, which takes time. Too many companies become reliant on this sort of revenue, and as a result, effectively transition into consulting firms, meaning they’re never able to finish their original projects. At best, this would significantly delay development. This was definitely not a solution for us.

While we put significant effort into the first two options, and some consideration into the second two options, all of them fell into some combination of costing us too much of our autonomy, not actually fixing our challenges, and burning too much of our time. So instead, we decided to take a measured risk:

Fund the rest of our development with a smaller project. As far as we saw it, there were two core difficulties with Scarlet Hollow from a marketing perspective:

1. It was hard to talk about.
2. It was unfinished, and it would at least be a couple of years, but probably several before we hit the finish line.

And at the same time, we identified a major strength:

• People really like the game once they play it.

And a corollary to that strength:

• People like our writing and Abby’s art.

Finally, we had one other important detail we could work with — our work schedule for Scarlet Hollow:

1. Abby writes the first draft of the next episode while Tony does post-episode release patches and marketing.
2. Tony writes the second draft of the next episode and codes it in the engine while Abby works on the art.

That first stage, specifically the Tony does marketing section, was a lot of time, only becoming longer with each subsequent episode, with very little real payoff. One can only post so much, and there are only so many marketing avenues one can pursue before exhausting them.

Therefore, what made sense to us was to develop a second game that we could work on in conjunction with Scarlet Hollow, that would fill Tony’s marketing time while minimizing its impact on Abby’s workflow. So, a game where:

• I write the first draft while Abby writes the first draft of the next episode of Scarlet Hollow. No need for elaboration here. This went smoothly, and by the time the script for Slay the Princess was done, 200,000 words of the script for Episode 5 were also done. Were Episode 5 not twice the size of Episode 4, this would have meant the WHOLE script would have been done by the time Slay the Princess’ script was done. But alas, we really want to make it feel like your choices matter, and Episode 5 is gonna kick ass in that department.
• With an art style that takes less time for Abby to draw than Scarlet Hollow’s style. For Slay the Princess, we moved away from Scarlet Hollow’s 18"x24" inch paper to 11"x17" inch paper, meaning each background had 56% less area Abby needed to cover. We also moved from pencils -> inks -> colors to just pencils (and then light colors), eliminating an entire step of our process.

Pictured: an inked background for Scarlet Hollow (left); a penciled background for Slay the Princess (right); (possibly) human hand for scale.

• And a smaller scope. Here, we would constrain ourselves to only having one character other than the player present on screen at a time, and would restrict the game to a single location.

“Yeah, so the whole game just takes place in one cabin so you don’t have to draw a done of different backgrounds.” — Me, expressing my hubris. We are nothing if not ambitious.

• A strong hook that can be conveyed in a quick sentence.

This hook proved to be VERY effective.

• Since the game has a smaller scope, we can add voice acting to make it more accessible. We were fortunate enough that our top picks for each character were immediately interested, and they both did a fantastic job. Thank you to Jonny and Nichole!

I’d like to note at the end of this list that some folks have seen these constraints and erroneously assumed that Slay the Princess somehow wasn’t a passion project. I’m a firm believer that constraints lead to interesting art, and I would hope that the execution of the game makes it clear how much we care about it! We love both our children equally, and we really mean it.

Black Tabby Games can have two passion projects, as a treat.

Before we announced anything, we wanted to put our theory about Slay the Princess’ development cycle to the test by working on a demo behind closed doors and seeing how long it would take us to finish it. The demo took me two weeks of work for the writing, coding, and voice recording, and it took Abby one week of work for the art, well within our constraints for a tight turnaround. We announced the game and released the demo in a one week window leading up to Tiny Teams, and that release was a resounding success, with Slay the Princess generating ~25,000 wishlists in two weeks.

If you want to read about our marketing strategy for that release in more detail, we have a handy-dandy post for you here: https://medium.com/p/3c69304312bc

Part of our hope with Slay the Princess was that we would be able to use a finished game to drive more interest directly towards Scarlet Hollow, circumventing the latter’s marketing challenges, and on this front, it was a resounding success. Announcing Slay the Princess and releasing its demo immediately gave Scarlet Hollow a 5x lift in its average daily sales, which settled down to a resting baseline of 3x its pre-Slay the Princess average.

As we mentioned earlier, 80% of Scarlet Hollow’s sales happened after this announcement, and over 33% of Scarlet Hollow’s sales have happened after Slay the Princess’ release, even though that was a scant 8 months ago.

Not only did Slay the Princess generate enough revenue to fund our studio for the foreseeable future, including for a third game, but it also elevated Scarlet Hollow’s profile to the point where, following Slay the Princess’ release, Scarlet Hollow is now profitable on its own, all while still in Early Access.

And we’re extremely happy with how Slay the Princess has performed overall, with a 90 Metacritic score (placing it in the top 5 PC game released last year, one of the most competitive years in game development) and multiple awards under its belt. And we’re extremely excited to share an even better version of Slay the Princess with you later this year!

So where does all of this leave us?

For those of you who have been with us since the start of Scarlet Hollow’s development, you’d know that every time we’ve released a new episode, a massive patch (or a series of patches) would follow in its wake as we finally had a chance to catch our breath, reflect on our work, and see what parts of it were and weren’t connecting with people. With Episode One, for instance, this large patch introduced Talk to Animals, an entirely new trait that greatly expanded the game’s replayability and breathed life into some of the community’s favorite characters.

Where would we be without Dustin and his mother’s sage advice? Probably dead.

The Pristine Cut is that patch for Slay the Princess — expanding on everything that we wanted to expand on, tightening the story’s pacing, and adding new features to improve the overall experience, including a huge CG gallery to help players find new routes and endings.

Look how cute this gallery is!

You’ve gotten a taste of The Pristine Cut with The End of Everything Update, which overhauled the game’s ending, added animations, and revamped some of the game’s pivotal tracks with a full orchestra. We’re excited about the flexibility Slay the Princess’ success affords us for Scarlet Hollow, where we’ll now have more freedom to bring in session musicians for tracks that would be elevated by live performances. It’s helping to elevate Scarlet Hollow in ways we never would have been able to consider before!

While we can’t say with certainly if The Pristine Cut is the final major update Slay the Princess will ever receive, for the time being, our primary work on the project is done, and our focus has wholly returned to Scarlet Hollow. All-in-all, work on the release version of Slay the Princess took me about 10 months of full-time work, and it took Abby about 6 months of full-time work, and The Pristine Cut took each of us another three months or so (plus two months of serious burnout on the tail of Slay the Princess’ original release.) So all-in-all, it’s probably slowed Scarlet Hollow’s development down by about 9 months. That said, I think these are 9 months we’ll be able to claw back before the end of development by virtue of being secure in our funding, and no longer needing to pursue other avenues to keep our studio afloat.

The future!

Overall, here’s our rough roadmap moving forward:

1. The Pristine Cut releases this Fall (date TBD)
2. The first alpha build of Scarlet Hollow: Episode 5 becomes available to our testers on Patreon before the end of the calendar year.
3. Scarlet Hollow: Episode 5 releases at some point next year.
4. Scarlet Hollow: Episodes 6 and 7 release together when they’re done.
5. Abby finishes her long-on-hiatus webcomic, The Last Halloween.
6. We start working on our third game. Some of this is probably going to happen during #5, since I’ll need something to do then :)

While we wanted to release the whole of Episode 5 this year, its increase in scope relative to Episode 4 is too much for us to set that goal without exposing us to grueling hours and severe burnout. Previously, that burnout was an unavoidable side effect of keeping the lights on — we had to work 60–80 hours a week, every week to release the next update in time to survive. Now that we’re not in danger of running out of funding, we’ll be able to move at a more sustainable pace, which will be much better not only for the rest of Scarlet Hollow’s development, but also for the development of future projects. In our quest to stay afloat, we’ve also been putting a lot of major life events on hold, including having a proper wedding (though that was more of a Covid thing) and moving into a bigger space so the two of us aren’t working/living/doing everything in a 600 square foot apartment, which is pretty difficult when your work includes traditional art, which takes up a lot of space! Our storage shelves had been getting pretty crowded, it was only a matter of time before most of our space was just storage for massive sheets of paper.

Anyways, that about wraps things up — thank you all so much for reading this and for supporting us on this adventure. We’re really happy with how far we’ve been able to take things, and you’re all the only reason we’ve been able to make it here. We are so, so grateful to be able to continue as an independent studio, free to make the creative decisions we want to make without fear of running up against impossible deadlines. We can’t wait until we’re able to share more of Episode 5 with you!

Until then, the gang says hi!

Look at these cute designs!

All the best,
Tony and Abby