Really, what is my job? I am not a clairvoyant who can predicts what Steam will do or have some special intuition that gives you wishlists. All I do is I go to the hundreds of devs who are willing to share their data with me (like you) and I ask them “did this work?” or “What happened here?”. Or I compile all the publicly available data filtered for what I am looking to test.

So what did I learn about Steam this year?

#1 Who should do Early Access?

The big question that is DMed to me or asked in the discord is “Should we do early access?” also “Is our game right for Early Access?”

This year I worked with the folks at VGInsights to see which games had the biggest improvements from EA launch vs 1.0 launch. I also talked to a bunch of developers who are VERY good at Early Access to see what they think. 

Here is who should do Early Access:

  1. You have gathered at least 7,000 wishlists (your EA launch IS your launch).
  2. You have spent at least 6 months marketing the game under a coming soon page (again, your Launch is your Launch)
  3. You are making a Crafty-buildy-simulationy-management-strategy-game (definition) with endlessly replayable gameplay.
  4. You have about 10-20 hours of content ready for people to consume.
  5. You know the game is pretty bug free after testing it via betas, playtests, and/or public Demos. 
  6. The game is fun. Don’t release a bad game. Only good games that are fun should do Early Access. 
  7. Your team and your development practices are setup for continuous development where you release an update at a regular cadence never longer than one month. 
  8. You love community building and constantly communicating with your players about what you are hearing from them and what you will be doing for them in the next patch.

From all my interviews, and analysis, the main reason you do Early Access is because you have designed a very good core system that is fun and replayable but only for about 10-20 hours. You don’t know how to add another 20-30 of hours of content on top of that without getting user feedback from Early access. Early Access is needed because only 1000s of players destroying your late game 24-hours a day can give you enough data to make your game fun for super fans. Steam players love LOOOOONG games. EA is to make your game longer so people don’t leave reviews like “I think the developers ran out of ideas after the first 5 hours.” 

Here is the risk of Early Access that I don’t think get considered enough:

  1. Players are skeptical of games launching into Early Access. They have been burned too many times by game developers disappearing after the initial EA launch and never ever patching the game and never going to 1.0. 
  2. This “player skepticism” leads to a potential lower conversion rate unless your game is SUPER HOT. 
  3. If your game does not sell well in the first month, there is little you can do to resurrect it. See the chart below. 
  4. Meanwhile, the players who did buy the game are very demanding. So if you have a game that didn’t sell well and you are not sure it is worth your time to continue developing it, fans will be  screaming at you “When is the next update! Dead game! Dead game!” 
  5. You can’t just abandon them otherwise your company gets known for releasing unsupported games. 

Simon Carless recently published a study that found that the median game 1.0 launch sells about .70x of the EA launch.

In this chart I scraped data from every Steam game that has every done both a Early Access and 1.0 launch. I found that if a game earned fewer than 100 reviews after 1 month of EA, they have a 5% chance of actually getting to 1000 total reviews after one month of 1.0 launch.

Don’t do early access if any of these are true:

  1. You are out of money and your game isn’t done so you  hope that by doing EA you can earn enough money to get the game over the finish line. This is trying to draw to an inside straight. Game development is very very risky and chances are your game won’t sell enough to continue development. So you will be left with not enough money to continued development and many angry fans screaming “Dead Game.” 
  2. No, the 1.0 release will not magically be better than your EA launch.
  3. Seriously, if your game underperformed at EA launch, I am sorry, there really aren’t that many cases of games “going viral” and turning around a poor EA launch. 
  4. You are making a linear game where once someone plays it, they don’t want to play it again. And no, adding a level editor or a speed running leaderboard will not make enough people want to replay it. The game must be a crafty-buildy style game.

For more details on these finding here are all the blog posts:

#2 Demos are the best way to get visibility and Valve thinks so too.

The pandemic forced people to make demos for virtual shows and we discovered that it works very well. Using demos and festivals felt kind of like we were exploiting a loophole in the Steam visibility and it was bound to close at any time. Some developers even question whether demos were good for your game and could actually, secretly be costing you sales.

But nope, none of this turned out to be true. Valve did the opposite. It’s as if Valve said “we see you doing these cobbled together demo tricks to get visibility and we …. love it! Here we officially added these tools to our system, keep doing it!” 

This is like Constantine adopting Christianity. Valve supports demos!

So what did valve add this year to codify demos?

  1. You can now get front page featuring when you launch your demo.
  2. You can now notify people who wishlisted your game when you launch the demo.
  3. Players can now leave reviews on your demo

Look at this game CleanFall that got a huge boost in wishlists when they launched their demo and ended up getting front page featuring. (full blog here)

#3 Should we use the “separate page” feature?

When you launch a demo you have two options:

  1. Create a separate Steam page just for your demo which turns on this feature where people can leave a review for that demo.
  2. Attach the demo to your main page and nobody can review it.

So should you host the demo on a separate page?

Yes!

In my research I found there was a tiny tiny difference in wishlists earned between games with the demo on the same page and those on a separate page.

However, having a separate page allows you to get written reviews from players. While this might seem like a recipe for getting your ego crushed, I think it is good. You need feedback from players! If they say bad things, FIX IT! Better to learn these problems now than when you release the game. 

Read my research behind the separate demo Steam page.

#4 What does it take to get a Weekend deals?

The biggest way to promote your game post launch is to ask Valve to feature your game in this Special Offers section. 

The minimum requirement is to earn somewhere in the neighborhood of $150,000 and with that you can get a “Daily Deal.” 

It’s pretty good but if you really want to make money, you want the Weekend Deal. This is the big money maker. 

What is the minimum bar to get a Weekend Deal? I worked with Ichiro Lambe of Totally Human Media to scrape the Steam Front page every day and see what games how many reviews they had. 

What we found is that if you have a hot game that has sold enough to get 20K reviews fast, go pester Steam for a weekend deal. Do it. 

Median: 19,990 reviews

Average: 54,068 reviews

I know this is a very big number, and most people are just trying to sell enough to quit the day job, but it is worth seeing how high the bars are for these featurings. 

#5 Should you launch your game right after Steam Next Fest?

You put your game in Steam Next Fest, you earned a couple thousand wishlists, your game is feature complete and almost ready for release, should you launch it right after Steam Next Fest since you have all this “momentum.”

No

The reason: Everyone has this same idea. In the 3 days following Steam Next Fest, every indie that thinks they have “momentum” tries to release their game. This leads to an over-crowded launch where you are all fighting for the same 10 slots in the Popular Upcoming and New & Trending widget. 

Here is a graph showing how many games (via their hype) are launching each day following Next Fest.

The better bet, maybe, is to launch 5 or 6 days after Next Fest or during the start of the Seasonal Sale if one occurs after Next Fest. 

Those days have many fewer releases. 

#6 Narrative Crafty Buildy

The best selling genres on Steam are games built upon an infinitely replayable structure that I call “Crafty-buildy.” This year, I saw some really neat games embrace “Crafty-buildy” and truly innovate upon it and add more narrative elements. 

Studying the market isn’t about copying a hit game or “chasing trends.” Instead, it is about looking at what fans find interesting, seeing what features nobody has considered, and then giving them a new experience within those interests.  Following market trends does not mean you are surrendering creativity.

So if you say you aren’t motivated to make a game in the “crafty-buildy” genres because you don’t like those types of games and instead prefer more linear games, please try these 3 games. They are doing innovative, creative things, in genres that indies don’t typically make but players love. 

There are even free Demos available for them 

#7 People don’t really play your demo but that is ok

This year for games that participated in Steam Next Fest, Valve started reporting how many people played your demo but didn’t wishlist, how many wishlisted it but didn’t play it, and how many played and wishlisted. 

I asked developers to send me their stats (Read the full blog here) and generated this venn diagram based on the median game in Next Fest:

Even if I look at the top performing game that I surveyed, you can see that most people don’t play and then wishlist. It might not look like it, but this is peak performance on Steam when you launch your demo:

Here is how I interpret this. 

I have shown over and over that providing a demo for shoppers increases wishlists earned. But this chart shows that very few people actually play the demo then go wishlist it. Most people will just see a demo, and wishlist it without even trying it. 

So why even do a demo?

Because providing a demo gives Youtubers and Twitch streamers something to play which generates traffic, which triggers the Steam algorithm to show your game more, and the very act of seeing your game encourages people to wishlist. 

Wishlists don’t actually come from people playing it. There are second order effects. Steam works in mysterious ways.

Thank you for 2024

Thank you so much for reading my blog, buying my courses, and sending me your data. I appreciate you and looking forward to figuring out and sharing more insights next yar.