Early Access without Steam?

Crossposted from Sharkbomb Studios blog.

For most people Early Access is synonymous with Steam. But with me being a solo-developer and Steam being such a big and important marketplace, I did not want to start with my game there. So instead, I launched Nowhere Prophet on itch.io.

Now, 6 months later I can say it was the right decision. So far I’ve made more than $10,000 in gross revenue on and a noticeable amount of that is from voluntary tips and the kickstarter-like rewards I offer, things that would not have been possible on Steam.

However Steam still plays a role, even when you’re selling elsewhere. Nowhere Prophet has been on Steam ever since the death of greenlight and a visible chunk of traffic and purchases comes from Steam, but more importantly: The sales spikes on itch.io correlate with increased wishlist additions, a key factor for launch success on Steam.

So, read on to find out what I did and how it all shakes out. Including all the interesting numbers for Nowhere Prophet. And as a little bonus: I’ve got some sales data from other itch.io titles for comparison, so you can look forward to even more graphs!

The Basics

Nowhere Prophet is a roguelike deck-building game that I’ve started working on in May of 2014. It’s been a long and winding road but since the 10th of October 2017 the game has been available in what I’ve dubbed First Access on the indie game store itch.io.

<iframe title="Nowhere Prophet - First Access Trailer" src="//www.youtube.com/embed/yHhGYvU2PbY?enablejsapi=1&amp;origin=https%3A%2F%2Fwww.gamedeveloper.com" height="360px" width="100%" data-testid="iframe" loading="lazy" scrolling="auto" class="optanon-category-C0004 ot-vscat-C0004 " data-gtm-yt-inspected-91172384_163="true" id="199303265" data-gtm-yt-inspected-91172384_165="true" data-gtm-yt-inspected-113="true"></iframe>

It’s Early Access and Crowdfunding in one. That means you can buy and play the game while I’m still developing it. That’s the Early Access part. However you can also spend even more money than the $19.99 price tag to have something of yours (like your name, your face etc.) become a part of the game.

First Access Goals

I’ve decided to run a First Access phase for three key reasons:

1. My Happiness

I’m a solo developer. That means I’ve been working on Nowhere Prophet mostly by myself. Granted there are a bunch of people helping out here and there but all in all it’s me against the world. And I’ve been doing that for long enough that development had become draining. I just knew I had to make the game ready for actual players. I had to get it out there. To see how it is received.

2. Their Feedback

I want to make this game as good as I can and that means I need some help. Again, working on it by myself means I have only one constant point of view on the systems and content. There’s no back and forth within a team on the daily issues. And that’s not good enough. So another goal was to get some actually player feedback to inform the design going forward.

3. Our Community

And lastly I wanted to start building a community around the game. Get people involved and excited. After all it’s consistently been one of the most rewarding things to see folks play the game at conventions so a lively community sounds like fun. It’s also helpful for spreading the word and making the launch a success, I hope.

I wrote an article on the first few weeks of the First Access, which also explains my reasons for going with itch over Steam. But now, let’s get into the numbers.

Gross Revenue

Let’s start with the most obvious graph, the sales of the last six months.

Note that these amounts (and all others, unless otherwise specified) are gross revenue. That is they are the money paid by the buyers to itch, without any deductions in place yet. The money that ends up in my hands is less than that. I’ll go over how this breaks down in a bit, but first, let’s look at the graph:

As you can see there’s a soft upwards movement in the valleys over time (until recently), but clearly more interesting are those spikes that are in there.

Let’s do some detective work on those dates.

The first one is obviously the launch spike, right at the gate. The second spike coincides with SplatterCatGaming’s two YouTube Videos showing off Nowhere Prophet. The one towards the end is from one backer that purchased the 350$ THE TALE option. And then there’s the towering spike in the middle of the graph, and that one’s clearly from the excellent Rock Paper Shotgun preview. There also is a slightly elevated plateau from when Pladd started streaming the game on Twitch (correctly named Follower Death Simulator 2018 on his streams) but that’s hard to pick out in the graph above.

Breaking down the impact of these sales periods is hard to do precisely, but here’s my best guess.

Revenue Spikes

First let’s look at a stretch without any spikes: The 70 days from 4th November to 12th of January. During that time period the sales stayed more or less level.

And these 70 days give us an average of revenue per day ($23.76). That’s going to be our baseline. Next we look at the general spike duration (~8 days) and track the revenue for each of the spikes for that time. And that gets us:

And lastly, we can do something pretty useless, but funny: Compare the "attributable revenue" from that spike with the number of views, followers or subscriptions. Let’s take a look at that.

SplatterCatGaming (YouTube)

SplatterCatGaming has 440,000 subscribers and both videos have a total of 95,141 views (82,391 for the first plus 12,750 for the second). Of course these are the numbers for now, not the numbers right after the 8 day period, but this is more for fun than anything really useful.

Rock Paper Shotgun

To measure the "reach" of Rock Paper Shotgun I took a look at their Facebook page likes (98,505) and Twitter followers (191,000). Obviously these aren’t the best way to estimate the visits on the article but it’s something.

Pladd (Twitch)

Pladd has a total of 3,120 followers on twitch. That makes for a whopping:

Some more thoughts:

My game updates don’t seem to correlate with any spikes. I’ve heard from other teams that big updates can actually create a visible push in the sales numbers, even without coverage, but that is probably due to those team having good reach and more time to advertise their changes.

Also I think the active development is probably what contributes to that soft rise over time. However the recent drop off is harder to pin down - maybe I’ve reached saturation on itch?

Revenue by Region

Another thing I can do with the data is to look at how the purchases and revenue are spread across different countries. In total there’s purchases from 46 different countries.

It’s not surprising that the US makes up the bulk of the purchases. Germany comes second, probably because I’m a German developer and I’m well connected here. That’s followed by Great Britain, the Netherlands, Canada and then France.

If we split it by continent, with Canada and the US lumped into their own thing, the graph becomes a bit easier to read. And you can see, unsurprisingly that the majority of the sales come from North America and Europe.

Revenue by Reward

Alright, now for the key thing that’s different from regular First Access, and one of my main reasons for Itch: The Rewards!

Considering I invested about 2 days to design, plan, get feedback on the tier rewards, and about a day to set up the page, make the graphics, write the descriptions, we come out at $290.11 per day spent on the tiers. That’s decent.

But that does not include the time spent to communicate with backers nor the time spent working on the actual content. So this number again, is more fun than actually useful.

However what is maybe slightly more interesting:

I have reached out to all my 38 reward-level backers via e-mail, some of them multiple times, but so far the amount of replies (and thus the implementation and fulfillment) is just above 50%. That means almost half of all reward-tier backers never reached out to me with the details of their reward. This includes a few of the people who paid $80 for the THE FACE reward.

I’m still a bit baffled by this, but not overly surprised, since this seems to be the case for many kickstarters as well. One would think that this might be friends and family supporting the game but not wanting the reward but that doesn’t seem to be the case. All the non-responsive backers are e-mails I do not recognize.

Revenue by Source

And one final thing on the revenue: Itch.io allows buyers to tip the developer, by setting a custom price themselves. This means there’s three sources of revenue: The base price, the difference to a reward price an the tips. Here’s how those break down:

What I found surprising in that is the sheer amount of tips. I have made some experiments with donations in games in the past but with very little success. It feels as if the itch.io buyer community is very much pro indie and willing to support developers with a little extra, if they can. It’s much appreciated.

Steam Wishlists

Another interesting set of data is related to Steam. The Nowhere Prophet store page and the community page have been available on Steam since Mid 2017, about half a year prior to the Early Access.

During that time it had accumulated almost 500 wishlist additions without being very active there. That’s neat. But what’s even more interesting is if we take the wishlist additions (and deletions) during the First Access period and compare them to the itch.io purchases:

As you can see these two are linked! That means that when the game