Staying Afloat - Keeping Up Sales in Early Access

While there are about a bajillion articles floating around about how to get attention for, and sell your game in today’s market, after releasing our own game We Need To Go Deeper into Steam’s Early Access program to moderate success, I realized that there aren’t really that many articles about how to KEEP selling your game once it’s already out on the market (with the exception of Lars’ Doucet’s excellent but a bit out of date article on The Stegosaurus Tail).

For an Early Access game especially, this kind of constant stream of sales is vital to both the continued development of the game and to the growth of the game’s community, and after finding out for ourselves exactly what that process is like over the course of our first 9 months on the service, I figure I’d share our successes and failures in keeping We Need To Go Deeper (and ourselves) afloat.

The First Wave

For context, and to help illuminate the exact position we found ourselves in after launch, first I’m going to talk briefly about our launch itself. The first month We Need To Go Deeper hit Early Access, we were fortunate enough to get covered by a pretty wide variety of heavy-hitting sources. From Roosterteeth, to Yogscast, to Markiplier, to RockPaperShotgun. (Our actual launch strategy is covered in more detail in this video Post-Mortem.)

It’s fair to say we hit the ground running pretty hard, and this initial coverage no doubt affected our ability/influence to garner more coverage in the coming months. We were very much legitimized early on, helping us avoid the initial pessimistic attitude towards Early Access games, and helping ground us with some decent word-of-mouth marketing.

But of course, like anything, hype and excitement quickly fizzles out as people look towards the next big thing, and this initial large-scale attention was short-lived. It was up to us to keep the momentum going, and turn this attention into long-form sales. From here on out, we had to make our game into its own next-big-thing every month. And with only a handful of indie-level examples to go off of, we formed a basic plan for how we wanted to accomplish this.

Updates as Events

To start, we looked to the way other successful Early Access games kept their community engaged and their momentum alive over their life-cycle. Don’t Starve and Nuclear Throne in particular being some of the best examples of this done well. In Don’t Starve’s case, each update seemed to be treated and marketed as if they were self-contained packages of content in their own right, which felt like getting Free DLC, as opposed to patch notes, every update cycle.

Klei accomplished this by creating these brilliantly beautiful “posters” for their updates, along with catchy names, like one might name a new expansion. This helped the updates feel special and big, like they were events to be celebrated.

We decided to run with this method in marketing our own updates, and ensured time was scheduled out each month to draw custom artwork for each update, along with a catchy (and innuendo-laden) name to go with it. In fact, we decided to ensure EVERY kind of update/patch notes/beta test would have some kind of visual artwork to go with it, to keep our store page looking colorful, and always mildly enticing. Plain text was to be a cardinal sin.

Rather than “posters” we opted for a wider “capsule” size for our Update Artwork. The plan was to send title-less versions of each update’s artwork to Press and Youtubers, to be useable in video thumbnails or image links.

Nuclear Throne had done something similar, producing original artwork to go with every update, which in their case was EVERY WEEK. While we were too keen on keeping our sanity to go the weekly route, we were inspired by the effect that kind of consistency appeared to have on Nuclear Throne’s community growth and retention.

As such, we opted to promise our community a content update would come out every month on a specific day (the 8th, based on our release date) so they would always have something to look forward to each month, and could count on revisiting the game regularly for the new content. This also furthered the cause in making Updates feel like EVENTS instead of patch notes.

NOTE: We did realise VERY early into the first month however, that once-a-month events wouldn’t be enough to keep our existing community alive and engaged throughout the month without thinking we were dead, so we opted to include smaller-scale “Crate Updates” mid-month that would be less heavily advertised and simply serve as a way to keep the community engaged/playing. This more or less put us on a two-week update schedule, God help our souls.

Life Preservers

The second part of the plan involved actually reaching out to Press, Youtubers, Streamers, etc. each month with these updates, sending them the original artwork, gifs, and a summary of the new content, in order to encourage the creation of new videos, articles, and streams - our “Life Preservers” - to keep the game visible in the public eye. As long as audiences were being exposed to WNTGD content semi-regularly, we figured word of mouth would continue to encourage a steady stream of sales.

Flares

We realized early on that not all updates would be equally interesting to cover from a streaming/video/journalism perspective, so we made an effort to always include in each month’s development goals a “Flare” feature (as I’ll call it here) - something usually flashy, particularly unique and visual that would attract content creators to return to the game and cover it. We figured as long as we could always make room for a “Flare” feature, that neither development or marketing would have to suffer.

The Results: Sales and Retention

So how did these strategies end up working out? Well, pretty well or pretty terribly depending on how you want to look at it. While these updates were certainly never able to keep up the kind of sales momentum we had in the game’s first couple of months, which in hindsight most certainly had less to do with our updates and more to do with our proximity to “newness” as we continued to benefit from the initial wave of coverage through large outlets. With that said, we were able to keep our heads above water throughout the last 9 months, but let’s look at some graphs and see for ourselves, eh?

Sales:

Red = Updates, Orange = Crate Updates, White = Prominent Events

As you can see, while major events like Launch, Sales, and coverage from extremely big Youtubers make themselves pretty well known here, it’s a lot harder to pinpoint the exact effects our Updates had on sales. Markiplier’s return coverage in May there was indeed prompted by the previous month’s update (Fully Loaded), but given its obvious rarity we’ll call that the outlier here in showcasing what sort of effects our Updates have had.

Most other content updates don’t at first appear to have any major effects on the graph’s current course, and that’d be fairly accurate. Though with that said, what they seem to occasionally do a good job at is stabilizing the downward trend. As updates get coverage, they seem to start leveling things out a bit, capping our bottom-limit sales, though the ones that don’t get covered are far less effective (See: Milking It).

Simultaneous Player Activity:

Red = Updates, Orange = Crate Updates, White = Prominent Events

Looking at our Players Graph reveals a bit of a different story. Here, we can see that while Updates don’t cause much in the way of spikes for sales, they do cause players to take notice and pick up the game again, with each update generally followed by increases in player activity. Being a multiplayer-focused title, it’s been vital to keep these numbers as high as possible as to avoid the “Indie Multiplayer Curse,” in which a multiplayer game gets slammed for not having a player-base and thus continues to not have a player-base. At the very least we can see what should be obvious in being the central purpose of our updates, and that’s keeping our audience engaged, involved, and talking about our game, so as to not completely disappear from public eye.

Successes and Blunders

Ultimately it appeared how effective an update would be at increasing/stabilizing sales or player activity depended largely on how easily understood and accessible the Flare feature we were promoting was. (and the more Youtube thumbnail friendly it was in terms of visuals).

Pictured: Our first artwork created exclusively so it could be used in Youtube thumbnails, as we shipped out a new submarine in the middle of the month to help us recover from the lack of coverage we got from the update itself.

New submarines were by far the most consistent way to net new coverage, particularly thanks to how easy they were to cover. From the get-go, Youtubers and streamers had a new ship to show off and play with by virtue of selecting it at the start of the game, as opposed to some of our RNG content additions, which despite their potential intrigue (Expansive, diverse cave systems and a Time Traveler NPC being a few examples), were much more difficult to get coverage from thanks to their random nature, which never guaranteed Youtubers would be able to see something headline-worthy within their short playthrough. The same could be said for some late-game content as content they couldn’t reach due to difficulty spikes may as well not be there at all.

For this reason, it proved beneficial later on to provide unlock codes to content creators, to allow them to show off new content without necessarily having to grind gold to unlock said content. This allowed new outfits and loadout items to be covered more in-depth, with several videos highlighting the use of our Dynamite item propping up in the process. In a similar vein, difficulty options/custom game sliders also helped content creators better hone their creations on what they wanted to showcase about the game. Challenge-based runs would still exist by virtue of these things being optional, but light-hearted goofier crews were now able to get farther, and new videos of past updates that had simply been buried in late-game content started to spring up once we added these things.

These methods to make it easier to cover content are growing all the more important as we start to shift into the more dry, technical side of our development, and it gets harder to justify spending time on “Flare” features that might otherwise take time away from more important, but less flashy game features. Especially when on a team of 4, this stuff gets a lot rougher to prioritize. On that note, our latest update aimed to add such a feature in the form of AI Bots, which while it’s too early to tell whether it’s successful or not, definitely should make coverage a lot easier in the future, by allowing empty player slots to be filled by Bots, removing the need to organize 4-party sessions.

Conclusion

I hope sharing our own experiences with this will help out anyone else struggling to keep their game alive during the Early Access phase. Some of this stuff seems pretty obvious in retrospect, but having had to stumble into these things through experience, I can say without hesitation that people aren’t kidding around when they say that marketing is a full-time gig.

The biggest takeaway throughout this process for me has simply been that all of the hard work and tactics that go into launching a game are the MINIMUM amount of work required to get people excited about your game again once it’s already launched. It only seems to get tougher the further out you are from release, so get out there and find those Life Preservers, because you can’t swim in the open water forever.