In part 1 of this post, we looked at the popularity of indie genres based on number of games and Steam owners. In this, the final part, we build on that to identify the genres presenting the best opportunity for indie developers.

Indie games by revenue

Revenue, simply put, is the number of owners of a game multiplied by the price they paid for it. However, owning a game – i.e. having it your Steam library – doesn’t mean that you paid full price for it. You may have picked it up at a discount, as part of a bundle, or even for free. There’s also the fact that we’re using estimates for the owners, so there will be a margin of error in those numbers too.

We opted to look initially at the maximum revenue a game would generate if every copy had been sold at the full price. Yes, this shoots for the moon, but we’re more interested in the relative comparison across genres rather than absolutes, and it’s easy to adjust the absolute numbers to ‘fairer’ estimates (as you’ll see in a bit!).

Maximum Revenue

We’ve kept the maximum revenue estimates in the order of most popular genres by game. The chart below demonstrates that being in a popular genre does not necessarily equate to maximum revenue opportunities. The “Action, Indie” genre still takes first place, but the maximum revenue on our second most popular game by genre type, “Casual, Indie” is much lower, coming 20th overall.

indie games max potential revenue

Reordering this same chart in order of maximum revenue shows that while action and adventure categories perform well, we see the emergence of RPG, simulation and strategy components in the top 10. “Action, Adventure, Indie, RPG” ranks 3rd and combinations of simulation, strategy and RPG appear in positions 5 to 10.

indie games max potentila revenue 2

We examined prices to understand how much these are a factor in maximum revenue estimates. The prices by genre are below. One clear take-away is that casual games carry a lower average initial price, with “casual, indie” having an average initial price of £3.43. This explains why casual games have a lower maximum revenue on average versus other genres with high numbers of owners. Games with RPG, simulation or strategy genres carry a slightly higher average initial price.

indie games initial price

Best Opportunities?

By best opportunity, we want to identify a genre, or set of genres, where they have a high number of owners, good maximum revenue potential, but are not in a space overcrowded by games. With the exception of the casual genre, price does not appear to be a significant factor in maximum sales estimates, so we largely discount this aspect. Instead we focus on a measure of average maximum revenue per game and distribution of revenue within each genre – is it concentrated in a few big titles or more evenly distributed?

Looking at average maximum revenue per game, it is the “indie, simulation” game, followed by “action, adventure, casual, indie, RPG” and “adventure, indie, RPG, strategy” that come out on top. Games including RPG, simulation and strategy coupled with either action or adventure genres make up 80% of the top 10 genres. The overall revenue per game is relatively high, ranging from £4.5m to 1.6m for the top 10. At the other end of the scale, casual and sports games have the lowest maximum revenue per game (around £55,000).

indie games max revenue per game

These numbers may be skewed by a small number of out-performers, so the below table shows the % of revenue accounted for by the top 5%, 10% and 25% of games by revenue in each genre. It also shows the average revenue per game excluding the top 5%, 10%, and 25% per genre.

indie games ave revenue

This reveals the concentration of revenue. In the “indie, simulation” genre, while it holds the highest average revenue across all games, the top 25% of games by revenue account for 98.3% of all revenue in the genre. With such a high concentration, the average revenue per game for those outside the top 25% is £105,321 – a mere shadow of the headline average of £4.5m

Concentration is relatively high across all genres, with the top 25% accounting for at least 93.3% of total genre revenue, meaning the average revenue of the games outside this sub-group is far lower. In the case of “action, adventure, indie, simulation”, this is as low as £71,570 per game.

Some genres do have a lower concentration than others. The “adventure, indie, RPG, strategy” genre has a concentration of 60.9% in the top 5% and 76.2% in the top 10%, but concentration at 25% is still high at 96.2%. That said, the average revenue outside the top 25% is relatively high at £191,229.

The “indie, simulation, strategy” has a concentration of 67.9% in the top 5%, 80.1% in the top 10% and 93.8% in the top 25%. This genre has the highest average revenue outside the top 25% at £199,779.

The “Indie, RPG, strategy” genre has a concentration of 64% in the top 5%, 77.1% in the top 10% and 93.3% in the top 25%. The genre has the third highest average revenue outside the top 25% at £143,648.

These three low concentration genres represent the best opportunity from the point of view that breaking into the top 25% of games in a genre is unlikely for most of us, and they carry the highest average revenue outside this segment.

Wait a minute…

We said earlier that the revenue estimates are the maximum revenue we believe could have been attained in perfect conditions (i.e. all owners bough their games at full initial price). The actual sales figures are clearly going to be lower once the effects of discounted or free sales are considered.

How much lower is unknown. We took the view a fairer estimate of actual revenue would be 50% of maximum potential, and the resultant revenue estimates for our three ‘preferred’ genres are as below. We’ve also shown the same statistics for the full sample as a control to help show the preferred genres have a far higher average revenue than the full sample set, regardless of whether we consider the maximum potential or the ‘fairer’ revenue estimates.

indie games top genres


From our analysis, taking into account the limitations of the data and the concentration of revenue in more successful titles, three genres stand out: ‘indie, simulation, strategy’, ‘adventure, indie, RPG, strategy’ and ‘indie, RPG, strategy’. These look like they present the best revenue opportunity for indie developers. If you’re lucky enough to want to make these games and have the time, budget and technical skills to do so, then you’re in a pretty good place.

For Sepia Cowboys, we’re not too sure that we fit that mould. But we have used some of our findings in our market research for our next project. More on that in due course!


EGX 2019
 My EGX wristband!

At some point in your early to mid 30s, hopefully without life altering events, you learn drinking is rubbish. It’s expensive, makes you an idiot and hurts like a bastard the next morning.

You eventually know this.

It’s just, sometimes, you forget.

On Wednesday I travelled south to stay with Dave before heading to EGX the next day. This was exciting for a number of reasons. Despite a lifelong passion for video games of all shapes and sizes I’ve never been to a ‘proper’ video game show. It was genuinely a bucket list thing. Also we’re *almost*, technically, in the strictest sense, game developers now. At some point we want to attend a show, exhibiting a video game we’re working on. It’s an opportunity to get a snapshot of the video game world and, sorry for using this word, network. Make some contacts, get some pointers as we start out.

Months ago when booking our tickets Dave suggested we should get the 10am early entry tickets. Make a day of it. Get all the benefit we can (we’re only there for one day). I don’t like travelling around London but Dave being a native will get me through it quickly. It will be fine. We’ll have plenty of time. We book the early tickets. We discuss the talks we want to see, games we want to check out and questions we should ask developers. We print business cards. We are prepared.

I look forward to this for months. Months.

Fast forward to the night before, it’s approximately 3am and we’ve had wine. Lot’s of wine. I’m not sure if it’s three or four bottles but we have long crossed the sensible-old-friends meeting-up-for-the-first-time-in-ages-and-having-a-drink line. A suggestion is made. Calvados! (French apple brandy). Seems a good idea. 

It’s a lovely drink.

At 4am a bolt of sense scorches through the air. We should get some sleep. 


Near EGX
We now live near EGX

At 8:15am either the world has exploded or it’s my phone alarm. 

Rise and shine!

One of us is, still, quite clearly, drunk and the other has passed that with a crushing hangover, a subhuman shell. It doesn’t matter who is who, we’re just two idiots on an adventure.

We need to change at Canning Town. We realise this as the train starts moving again. That’s fine we jump off at the next station and onto the train back to Canning Town. In a rush we dash off the train just reaching the next one before the doors close. Few. Made it. At this point I notice a person wearing a massive Pikachu hat. They didn’t get on this train. Off at the next station and onto the train back to Canning Town. We pause. Canning Town is beginning to feel a bit like home. We make damn well sure we get on the correct train. We hold our breath as the train pulls away. Success!

We get to EGX at 11:15am. The first thing that hits me is the size of the ExCeL. It is huuuuuuge. The entrance to EGX is on the opposite side from where we came in so we have a bit of a trek. One of us is bouncing along, the other shambling, drifting on a memory of humanity. Twenty minutes in the queue and we’re in.

EGX is everything I hoped it would be and more. It’s video game magic land and I’m here to be part of it! Except no, I’m not really myself, nowhere near.

We do manage to get a lot of benefit from EGX. We see how things are presented, we attend interesting talks. We see a lot of awesome looking indie games. I even get a quick glance at the new Avengers game. Ok, it’s not why we’re here but come on! 

It feels great just to be here.

But everything is happening slowly, the only pace we can manage. We speak to few developers. We don’t play as many games as we could have. We don’t go to after hours. We get tired.

We will look on the positives and we will do better next time.


Probably meet at Canning Town and head in from there.



One of the things I’ve learned from my project management life is that working on only one project at a time leads quickly to lulls in activity, where resource isn’t utilised fully. In fact, earlier this week I was talking to someone who ran their own indie games studio for 10 years and they said this was their biggest challenge – making sure there was a sufficient pipeline of ideas and projects to keep their dozens of staff productive.

The Sepia Cowboys may already have fallen foul of this already (eek!), so while we await the release of Hope, we’ve been thinking pretty hard about what to do next. I read a while ago that as developers there are (i) games people buy (ii) games we would like to make and (iii) games that we are able to make. At the intersection of these three groups, therefore, are those games that we probably should make. Being commercially minded, we thought a good starting point would be to understand what type of indie game sells, which is what we set out to do.

Data sources, limitations, assumptions

An immediate issue we discovered is the lack of really good data on video game sales. SteamSpy is a great source of current data on games and genres, but their estimates on owners can be quite wide, particularly for smaller indie games. There’s also the fabulous data gathered from the 16-decimal-place Steam achievement ‘glitch’, which produced pretty accurate owner numbers, but is now over a year old and only includes games with Steam achievements. We settled on using a combination of both sources – while not perfect, it was the best we had to go on.

We removed some games from our dataset – only indie games were included, free to play were out (controversial?) and gone were the non-core genres such as utilities. From an initial dataset of 13,000 games, this left us with a touch over 9,000 in our sample.

Indie games by genre

The first thing we looked at was the number of games in each genre. Games don’t only come with one genre, so we had the joy of over 370 unique combinations of genre tag in our sample. To make this more manageable we filtered this to only show those with 25 or more games, leaving us with 49 unique genre combinations. Perhaps not surprisingly, the most frequent genre is “action, indie” with 1,245 games, 12.5% of the sample. The next most popular genres are “casual, indie” with 813 games and “action, adventure, indie” with 802 games. Collectively, genre combinations of “action”, “adventure”, “indie” and “casual” account for 62% of the games in our sample.

indie games by genre

Looking at 49 bars is still quite a lot, so we then limited the number of genres to ‘main’ genres of action, adventure, casual, MMO, racing, RPG, simulation, sports and strategy and counted the number of times each of this was included in a genre combination. As games can have one of more of these genres, they may appear in multiple genre classes (i.e. the sum of the genres adds up to more than the number of games in our sample). The genre with most games is action, closely followed by adventure and casual, in line with the analysis on the fuller genre list. Middle of the road are the strategy, RPG and simulation genres. The genres with the fewest games are sports, racing and MMO.

indie games by main genre

Indie games by owner

Ok, but which games do people actually own? Now it gets a bit tricky. We have the SteamSpy source of owners that is current, but can be low on the accuracy front. We have the achievement source that’s out of date, but much more accurate. We shot for somewhere in between and generated a blended owners count, using a weighted average with double weight given to the achievement source.

indie games by owners

The “action, indie” genre had the most owners, followed by the “action, adventure, indie” genre. We see the emergence of “RPG” in third place, in the “action, adventure, indie, RPG” genre. While the top 10 – accounting for 65% of the owners in the sample – includes RPG, strategy and simulation elements, the action, adventure, indie and casual elements would appear to draw the most owners. At the opposite end, “casual, indie, sports” games fare the worst, with the fewest owners.

Conclusions & Next Steps

In terms of popularity – by number of games and owner count – action, adventure and casual indie games come out on top. Elements of RPG, strategy and simulation start to sneak in, while the sports, racing and MMO genres are decidedly niche. I can’t say there’s anything particularly surprising here.

Where it does get more interesting is where we go deeper and look at the revenue potential in each of these genres, to give pointers on which genres present the best opportunity for a small studio to pursue. And that’s exactly what we’ll do in part 2 in a few weeks… get your number brains warmed up!!!


It’s done! We have our Hope Coming Soon page up on Steam and are a short few weeks away from release. While it’s a milestone that’s been a long time coming, it’s now that we start to get a sense of how much the game strikes a chord with the Steam audience.

Wish(list)ing our lives away…

In the pre-release phase on Steam, the best indicator of likely sales for a title is the number of wishlists it has received. Anyone can go to your store page, click on the wishlist button, and add to the tiny glimmer of hope (no pun intended) that your game will be a success. After the launch of our Hope Coming Soon page, we pretty quickly got an idea of where those wishlists may come from.

Steamy Windows

Steam provides data on impressions – the number of times a capsule image of your game has been viewed – and the number of visits to your store home page. The idea being that impressions turn to visits, turn to wishlists, turn to win.

By far the best source of impressions was the Coming Soon list, delivering 30% of total impressions. As soon as a game enters Coming Soon, it is added to this list, so it should give continued impressions until release. However, we found it to be a pretty poor source of actual visits, with just over 1.4% of impressions turning to visits.

Second on the list was direct search results. This is where people have searched for a game, are browsing the results, and Hope has been displayed there. Just under 20% of total impressions came through this channel, and again, we saw a pretty low conversion to visits of 1.3%.

The next four sources brought us around 10% of total impressions each, with varying degrees of success. Individual users Steam home pages (i.e. discovery queue, main cluster, curator recommendations) saw a 9% conversion into visits. Steam curator or other developer home pages saw an 18% conversion to visits. Search suggestions (where Hope has been displayed as an auto-complete option) brought us an 11% conversion to visits, while people browsing results on Steam lists delivered just over a 2% conversion rate.

Overall, while these sources delivered a high number of impressions, they account for only 6% of total store page visits. They also saw higher numbers of impressions in the first few days of Coming Soon, but quickly dropped off over the next few days. It unlikely, therefore, that they’ll be much use over the next few weeks to release.

So where are those elusive store page visits, and hopefully, wishlists, coming from?

steam me up

Show me the visits!

We’ll it looks like there’s a big bucket of ‘direct navigation’ accounting for 32% of them. These are visits to our store page from a browser, where it’s not possible for Steam to identify the source. Quite a few sites do this automatically when an external link is given. The privacy and security conscious side of me says this is a good thing, while the indie game developer isn’t so sure. I guess what we do know is that these are non Steam sources.

The other great thing about direct navigation is that it appears to be a nice, relatively constant, source of visits. In the last 2 weeks it’s dropped off a little, but it’s still bringing in a nice, steady number of visits every day.

Another biggie on the visit list is personalised discovery queues, at 31% of our visits. So there are people out there with zombie pigeon games in their preferences. Wow. Rule 34 suddenly springs to mind too.

Finally, there are around 20% of visits coming from external sites – including Valve – but also the standard social media crowd.

Sorry, did I nod off there?

Possibly. We looked at these numbers to find out what was driving traffic to our store page. Our take on the above is that while 40% of store page visits may be down to Steam itself (and some pigeon fanciers) the majority is down to your own efforts at building your profile.

For me, this is encouraging, particularly with some of the recent changes Steam has made and their apparent negative effect on indie game developers.

With Hope we started late, but in the past few months we’ve built a website and started posting regularly on social media. We’ll aim to maintain that presence to release and beyond.

Oh, and if you’ve read this far…check out our store page already, will you?

rhino sleep

Hope Steam trailer capsule

There comes a point in most indie game developers lives where they have to hit that big red button and get their game up on Steam. We hit this point a few months ago with Hope and wanted to share a few pointers on what we learned getting that initial store presence up.

It will probably take longer than you think

Well, it did for us, anyway! Before you get to the fun bits, you’re going to need your Steam account and banking information set up. We opted to sign up as a company, which probably takes a bit longer than doing it as an individual. Getting incorporated in the UK is quick (we used as is getting your US tax ID to avoid US withholding tax (just call them!), but getting our business banking setup took around 4 weeks. Sadly, you can’t start work on your store page until this is all complete. Another factor to consider is that Steam has a review processes of 2-5 days before a store page becomes live. There’s also a minimum of 2 weeks between Coming Soon and launching and a minimum of 30 days for the whole process. Do consider all of this if you’ve some deadlines you want to hit!

Steam Early Access and Coming Soon

Rather than jump straight to selling on your store page, you begin with an Early Access or Coming Soon page. We opted for coming soon, as we have a near finished product (is anything ever really finished?). If you’re much earlier in development and are looking for community engagement to shape your product then Early Access may be more for you. Bear in mind if you do take that route, while you may get input, you’ll also be expected to keep delivering too.

It’s easier than you think (mostly)

Once you’re past the account and banking nasties, things get a whole lot easier. Steam’s documentation is pretty good and the on screen tips guide you through the process. I’m not an expert on the visual assets side of things (that’s Paul’s bag) but it was even easy for me to understand what was needed. And there are a lot of handy bits like auto-resizing of images for screenshots. Getting your build up though, is a bit fiddly. If you make an error when putting your build together, you’ll need to go right back to the beginning of the process rather than tweak your build.

A trailer is a must

These days, you’re not getting anywhere without a trailer. First, you’ll need to consider what you’re going to put in your trailer. You can get some great tips on this here, but in summary: tell a story, show off your gameplay and keep it under 90 seconds. The next thing is how you’re going to put it together. We used OBS for our video capture then photoshop to do the editing. There are many free editing programs out there if you don’t have access to photoshop. Most importantly though, your trailer is going to be the main reason people want to know more or hopefully even buy your game, so make it as good as it can be.

Do it early

With Hope, we’ve set up our Steam page pretty close to our expected release date. On our next project, we’d do it much sooner. The longer you have your Steam presence up, the longer you have to build interest, community, followers and get people adding your game to the fabled wishlist. People will wait a long time for a game they really want (HL3, anyone?) so there isn’t really a downside to getting your page up early.


There is a wealth of stuff that you can add to your store page over and above the basics, but I think I’ve covered enough for now.

Our Coming Soon page has been submitted and is undergoing review; so hopefully we’re only days away from getting our store presence up. Watch this space!

Hi, I’m Dalvor, the Sepia Cowboy that has a nickname. I’ve decided to start our devlog with a quick intro on who we are and what we’re up to. So here goes.

It all started around 2 years ago when Paul (the other Cowboy, with no nickname) decided making a video game would be fun. Ha! How little he know. He worked long days, late into the night, to churn out Hope as his first complete game.

That’s where I came along. Seeing that most of hard work was done, I convinced him to let me jump on the bandwagon and Sepia Cowboys was born. 

So here we are…our first game release only a short time away and some big ideas on the next one. As we choose our next project and start to work on it we’ll use this devlog to keep you up to speed… from early concept through to release and beyond. The good, the bad and the downright frustrating.