Which matters more: The game or the marketing?
Popular gamedev YouTube personality Jonas Tyroller recently released a video where he argued that the most important factor in your game’s success is whether the game is “good” and that “marketing” is not as important.
His argument is that if you have a “good game” the Steam algorithm will biggest source of traffic. Although marketing is still important to kick-start the process, it is becoming less of a factor in your success overall.
Here is the most important quote from the video
“Marketing is used very convenient excuse [for why your game underperformed] by Indie developers but NO, very likely it is not why your game failed…Algorithms are the new marketing.”
Jonas got some flack for this statement and he asked if I wanted to weigh in.
I do want to weigh in. But first, I looked at the data for his three Steam games to try and give him a firm answer. I don’t just trust my gut and make stuff up.
TL;DR: He is right. The type of game and its quality is the most important factor and most of the traffic that converts to sales will come from the Steam Store Algorithms. Let me spend a couple thousand words explaining how.
The 3 Games of his career
Jonas has built an enviable game dev career. He is constantly building, constantly releasing (he documented a challenge where he made 50 games in 1 day.). This guy ships.
But he doesn’t just release game-jam quality quick hit games. Since 2019 he (and collaborators) have released 3 full-sized games to Steam and all of them have been a “success” in that they made more money than they cost and they allowed him to continue to make games full time.
The 3 games vary in scope, genre, cost, and team size. However since Jonas has been a creative force between all 3 and done fairly similar marketing approaches, we can compare their relative performance and glean some insights. Jonas is about as close as a control in an experiment as you can get in this creative and chaotic industry.
So here are his three games:
- ISLANDERS (Released 2019)
- Will You Snail? (Released 2022) (WYS for the rest of the blog)
- Thronefall (Released 2023)
Now because of various partnerships (and the fact that he sold the IP of ISLANDERS) we can’t share all the numbers. But Jonas provided a high level summation of how they performed in the form of this table. Jonas also created a “Subjective Success Score” that sums up how the game performed based on all these metrics, roughly reflecting the financial success of each game so that we can see how they relate to each other.
ISLANDERS | Will You Snail? | Thronefall |
| Released 2019 | Released 2022 | Released 2023 |
| Traditional full release | Traditional full release | Early Access (still no 1.0) |
| ~20K Wishlists at Launch | ~60K Wishlists at Launch | ~200K Wishlists at Launch |
| ~1500 CCU on Launch with a decent tail | ~700 CCU on Launch but the tail died off very quickly | ~7000 CCU on Launch with a great tail so far (probably due to the regular updates during early access) |
| ~15K Reviews in 5 Years | ~2.3K Reviews in 3 Years | ~15K Reviews in 1 Year |
| Sells for ~$5 USD | Sells for ~$12 USD | Sells for ~$12 USD (but was much cheaper at the beginning of Early Access) |
| A fairly large player base but not a super passionate one and not one that stuck around for very long. Average playtime for the game was surprisingly low, but people liked it. | A much smaller community, but a surprisingly passionate one. It’s not for everyone, but those who like it really do. | For the first time tapping into that sweet spot of both having lots of players, but also many who are quite passionate about the game (not as crazy as WYS but noticeably more so than ISLANDERS). |
| Content Creator coverage on YouTube and Twitch was good but very short lived. | Content Creator coverage on YouTube and Twitch was rather sparse. | Content Creator coverage on YouTube and Twitch was great with many creators covering the game multiple times, some even for every new update. |
| More than paid for itself! | I’d say it roughly paid for itself and then some (including a decent salary). | More than paid for itself! Sold similar numbers to ISLANDERS, but in a shorter time and with a higher price point. |
| Subjective Success Score: 50 | Subjective Success Score: 15 | Subjective Success Score: 100 |
The general rule of thumb SALES = REVIEWS x 50 also works decently well for estimating rough ballpark numbers for the number of units that were sold (as I’m sure you’re aware).
Chris Analysis
I have been following Jonas’s career since the amazing launch of ISLANDERS. I thought ISLANDERS was the perfect type of game indie developers should make:
- The genre is “buildy” – Your primary verb is placing buildings
- It takes a complex genre (City builder) and simplifies it to its essence so that it appeals to the fans but keeps development times low (6 month development time)
- The game design is endlessly replayable. There is no narrative or puzzles or “levels” so fans can play it over and over.
His second game was a platformer…
If you have been following my writing and talks for a while you will know that I strongly recommend that developers avoid the platformer genre for many reasons.
But WYS did great! It was a hit! It has over 2500 reviews!
“See Chris! Some platformers do well!” – That guy on reddit
Yes, WYS did great! It did better than I thought it would. But, even a top performing platformer will still underperform compared to Jonas’s other games in the genres that Steam players prefer: “Crafty-Buildy-Strategy-Simulation” games. Jonas is a very experienced game designer, with 200,000+ youtube subscribers, and even with all that, he found it a challenge to get a platformer to succeed.
WYS probably got a bigger hokeystick than it deserved to at the beginning of its life cycle due to my YouTube channel and pretty decent wishlist counts.
Jonas
Let’s look at it in the next section.
Concurrent players
I graphed the peak concurrent players (as provided by VGInsights.com) for Jonas’s 3 games.
Yes Thronefall’s traffic dwarfs his other 2 games. But if you look closely at the red WYS line you will notice that the traffic falls off quickly and there are no post-launch spikes. Notice how both ISLANDERS and Thronefall enjoyed big sales spikes many weeks after launch.
I think the truth is, [Will You Snail?] is not as good as the other two releases of mine, especially when it comes to appeal. The September 23 spike [barely visible on the WYS line above] came from the Level Editor update. I was hoping this would infuse some new life into the game and give it a much better tail, but as you can see it absolutely did not achieve that at all (the active part of the community was still very happy about it).
For ISLANDERS the spikes can only come from sales because we never updated the game with any significant amounts of content (besides the creative mode and an undo button, but that was relatively shortly after launch). I think with low-priced games at 5 USD and below, the sales really help with impulse purchases, because the games are just very very cheap at that point.
Jonas
Also regarding Youtube and Twitch streamers playing WYS Jonas said “Content Creator coverage on YouTube and Twitch was rather sparse.“
So if you are making a platformer, and even if you do everything right and gather a pile of wishlists and convert them, they often fade quickly. Streamers don’t like to play platformers as much as other genres, and content updates and sales just don’t move the needle as much.
Solo development vs Team development
Jonas ranked his games based on a “Success Score” that let us compare the rough financial success of the games without having to know the exact numbers. Here they are again:
- ISLANDERS: 50
- Will You Snail?: 15
- Thronefall: 100
But I wondered, WYS was a solo game whereas ISLANDERS and Thronefall were team efforts where profits would have to be distributed with other people. Also WYS had a higher price than ISLANDERS. So which was the most profitable to him as an individual?
Jonas broke the data out for me based on team size and time worked on each game
| ISLANDERS | Will You Snail? | Thronefall |
| 3 Man Team | 1 Man Team | 2 Man Team |
| ~6 Months of Development | ~4 Years of Development | ~3 Years of Development |
| = ~1.5 “Manyears” of work | = ~4 “Manyears” of work | = ~6 “Manyears” of work |
He then divided his subjective success scores (see above) by the amount of years it took to make each game (see “Manyears” above) to get the “real value” of his games.
| “Real Value” = 50/1.5 = 33.3 | “Real Value” = 15/4 = 3.75 | “Real Value” = 100/6 = 16.6 |
I never made this calculation before. I must admit, I myself am kind of surprised by these results, but it seems about right. It almost feels somewhat sobering to see this now. You often forget to consider the time and resources you spent when looking at raw results.
Jonas
Jonas’s first game, which he made in a short 6 months, was his most profitable when you factor in time spent on it.
We made it in an incredibly short amount of time and it did amazingly well considering the resources we spent on making it. We made this as a ~3 month semester project as a part of our game design studies at HTW Berlin and spent another 3 months of our own time after that to get it into a release-ready state.
Jonas
By these numbers, his lowest performing game was WYS by quite a margin.
Will You Snail? was a passion project of mine and if I only made games like this it would probably be very very barely enough to make a living, though likely possible. I did not expect it to perform so poorly on the “real value” benchmark here for that reason. It felt like I was doing very okay and I think many developers would be super happy about a result like that. If anything this shows that results are not “fair” in the way that you’re not getting more out of it just because you put more in.
Maybe this is a good time to point out that “success” can be measured by many different metrics and financial success is certainly not the only one you should measure. For example Will You Snail gave me a lot of creative fulfillment and creative freedom that the other two games couldn’t live up to. It’s important to find a good middleground and work on things you enjoy! Create your own definition of “success” and optimize for the whole package!
Jonas
Chris Analysis
Again, I want to emphasize that I am not picking on WYS. It did very well by any measure. Also money isn’t the only marker of success – especially in a creative endeavor like game development.
If you just look at the number of reviews a game has you could conclude that WYS is one of those exceptions to the rule that platformers underperform. But when you dig deeper the “platformer tax” becomes more clear.
Discovery Queue Magic
I have written about what the Discovery Queue is in previous posts and how to get on the Discovery Queue pre-launch, but I cannot emphasize enough how important the Discovery Queue is for generating sales. Not enough developers are aware of it, not enough developers are concerned about it.
Discovery queue is by far the biggest source of traffic for both of these games in these launch days I looked at. Like, over 50%, often much more of the store page views come directly from the discovery queue. That’s insanity! I was not aware.
Jonas was able to give me Discovery Queue data for two of his games so I compared how they performed:
I have looked at a lot of post-launch Discovery Queue charts for many games and I see a consistent pattern:
Phase 1: Growth phase
- For games that gather a lot of wishlists they typically are given around 20,000 Discovery Queue impressions on the first day of the launch (Notice both of Jonas’s games got this).
- If the game “performs well” with that first 20,000, Steam shows the game in the DQ at least twice as much the next day. I don’t know what “performs well” means to the algorithm: it could be sales, it could be wishlists, it could be the number of times the game is not “ignored.”
- In the following days Steam gives more and more traffic (typically twice as much as the previous day) until the game stops “performing well” and then traffic is cut and the decay phase starts.
Phase 2: The decay phase
- Once the game peaks in the DQ, traffic is cut for each of the subsequent day
- Games that lasted longer in the DQ-growth phase seem to have a more gradual “decay phase” than games that peaked early
My theory is that in the first few days, Steam shows the game to shoppers that have the highest play times with the game’s assigned tags – they are the most likely audience for the game. Then in subsequent days, the Steam Algorithm shares the game to a broader audience that has a lower tag affinity to the game. This goes on and on until the game starts to “underperform” with an audience. Then traffic is pulled.
Chris Analysis
When Jonas said (and I agree with) that Steam does most of the promotion for you, we are talking about the Discovery Queue (DQ). The Discovery Queue is why you give 30% to Valve. You are paying for them to show your game to the purest, most hardcore fans who love to spend money on video games. If the game doesn’t perform well in the DQ, there is no way you can find a more dedicated, more spendthrift audience. But Valve controls the flow of the DQ depending on how “good” your game is.
If you look at the chart comparing Discovery Queue traffic for the two games you can see how much more Thronefall is featured.
WYS traffic grew for 2 days before it was cut. THAT IS GOOD. Most games don’t even get a second growth day in the Discovery Queue.
But even though WYS performed relatively well in the DQ, Thronefall is a more traditional “crafty-buildy-strategy-game” which has a much bigger audience on Steam. There are more people that the DQ can show the game to and there are more people willing to take a chance on it. Platformers just don’t have that same built-in audience on Steam (even if the game is top notch) and so it faces stiff headwinds.
If we assume that this transfers to other indie games, then having an appealing game that makes a great first impression when people come across it in the discovery queue, could indeed be much more true that I even imagined.
Jonas
Does Youtube Help?
Jonas has a huge and loyal Youtube following. But, is it worth the effort to grow a YouTube audience to market your game?
For Will You Snail, most definitely. I think most of my wishlists come directly from YouTube. However, it was a slow process. I worked on the game for 4 years with 4 years of devlogs on YouTube. If you imagine you can post one video and get 10K wishlists from it, you are probably mistaken. I’d pretty confidently claim there are more efficient ways to get wishlists than from running a YouTube channel (which is a huge huge time investment).
Look at Thronefall for example, here my own videos don’t even make a noticeable dent in the wishlist count at all.
That’s because WYS launched with somewhere around 40-60K wishlists while Thronefall was more at 200K or so and by now it’s grown even higher to 500K+. If I published a YouTube video asking people to wishlist Thronefall now, you wouldn’t see a damn thing on the wishlist graph.
Can a YouTube channel ensure a reasonable baseline of wishlists? Yes.
Does it help? Yes.
Can it blow the wishlist count out of the water? Unless you are Dani, no.Is a YouTube channel the most efficient way to get wishlists? Hell, no.
Look at it like this: Instead of building a YouTube channel yourself, use the channels of other creators who are already popular. If your game is good and interesting enough (and that means better than the other available options) they will likely play and feature your game, so you bypass all of the hard work of having to build that audience yourself. -> Very easy to argue that that is a much more efficient way to go about it and on top of that it gets your game in front of gamers instead of in front of game developers.
Jonas
Summary
So, to answer the question that Jonas put out there “Is the game more important than marketing?”
Unequivocally I say “Yes.”
(Side note, the game you choose actually is the marketing if you are familiar with the 4Ps of marketing and want to be extra pedantic about what is marketing).
I think in general it would be worth mentioning that “marketing” is not a super well defined term and a lot of the disagreements and arguments stem from having different definitions for what marketing is. You could also say that making a great game is marketing.
You see, the moment you pick what type of game you are going to make and that you are going to release it on Steam, you have made 90% of your marketing decisions right there.
From a purely “marketing” standpoint, I think Jonas’s strongest game was ISLANDERS. He made a crafty-buildy-simulationy-strategy game in under a year, built up a decent collection of wishlists and then released it, he was making the right product for the right audience and he came in way under budget. Even if you make the best game, at some point, if you spend too much time making it, it will become unprofitable.
When Jonas made WYS, he performed waaay above the median for a typical platformer. He deserves so much praise for that. It is a great game. However, the dynamics of the Steam audience mean that even if you do everything perfectly with your promotion (such as build up a huge YouTube following), you cannot overcome the downward force that is making a platformer on Steam.
That downward force is evident when you look at the concurrent user graphs being so much flatter than his non-platformer games. You can see it in the way he said that streamers (with some exceptions of course) just didn’t want to play a platformer. You can see it in the Discovery Queue where the algorithm saw that people just weren’t interested in it so it cut the traffic.
I know this blog is looking at just one developer with 3 games. I know this is not statistically relevant. But this is a qualitative study. I see similar results when I do a broad-based data scrape of every game released every year.
I see similar results when I just look at the top 1% of games from each genre.
I see it when I look at the top sellers every year.
Games that do not fit the genre preferences of the Steam audience underperform no matter how big the developer’s Youtube channel is, no matter how spicy their tweets are, no matter what crazy promotion you do. The game you pick is the biggest marketing decision you will make.
So if you are looking to make games professionally so that you can earn a more-sustainable income, I recommend making games in the genres Steam likes (crafty buildy) and don’t spend too long making them.
If the game you pick is the biggest marketing decision, then your game is your most important marketing asset. There is huge demand on Steam for small but highly-polished experiences. Put quality over quantity. Just make great games that people actually want to play. There is no better thing you can do for the marketing (however you define that) and the success of your game.
Jonas
Help demystify the Discovery Queue
If 50% of your launch visibility comes from the Discovery Queue we should definitely study it more right?
Well I am launching a new survey trying to see how different games performed in the discovery queue. If you have released a game (both Early Access or Full release) and both paid and free, please submit your data for my latest survey: