Steam is removing Greenlight – An Indie Developer’s Interpretation
Steam Cancels Greenlight
Valve has recently announced that they are planning to remove the Greenlight system that launched in 2012. As many already know, I am an indie game developer that put my game, Soul Saga, through the Greenlight submission system in 2013, just one year after Greenlight launched.
Soul Saga went through very smoothly within about a month’s time. I found the process to be very flattering knowing that so many people were interested in Soul Saga, and thought that the system seemed to work well from my experience. However, the Greenlight system is now set to be removed completely as per this news post by Valve:
The next step in these improvements is to establish a new direct sign-up system for developers to put their games on Steam. This new path, which we’re calling “Steam Direct,” is targeted for Spring 2017 and will replace Steam Greenlight.
There’s a lot of interesting comments regarding this rather important news of Steam’s submission process, but it seems many are focusing on single distorted details and misguided assumptions rather than the core messages delivered in the update. As an indie game developer that has successfully gone through the Greenlight submission process, I would like to give you a developer’s educated and experienced interpretation of this news. There’s a lot to parse here, so let’s keep it organized:
The Amount of Games on Steam
Many players have stated hope that this change will reduce the amount of “bad” games on Steam, and (in their assumptions) increase the percentage of “quality” games. This news post shows no interest in REDUCING the amount of games on Steam. It is consistently saying it wants to remove the barriers to entry, which would actually INCREASE the amount of games on Steam. So the people who think getting rid of Greenlight will stop the influx of (their words) “crappy games”, they are very wrong. Valve basically outlines their evolution of their submission process as the following:
- The Past – Manual game filtration by Valve EMPLOYEES. (Few games got through.)
- The Present – Crowd game filtration by Steam PLAYERS. (Many games got through.)
- The Future – NO SUBMISSION FILTRATION. (All games will get through!)
Many have considered the future of no submission filtering a bad thing. However Valve has stated here that they are not interested in telling people what they can and can’t buy. They are more interested in an open and free market, something that can be very positive (especially with refund policies). The only barrier to entry would be…
Pay to Publish
Valve has stated a (wide) potential range of $100 – $5,000 as an entry fee to publish on Steam.
Once set up, developers will pay a recoupable application fee for each new title they wish to distribute, which is intended to decrease the noise in the submission pipeline.
As a developer I’ve seen all too often how consumers’ jaws drop at the site of 0’s. I myself was in that position until I started developing games and investing money into my projects to pay freelancers, licenses, and fees for my project. I can tell you that this range, even at the highest point, is not unreasonable. Console development is thousands at the lowest end and gets more expensive the larger the team becomes. Some clarifying points:
Point 1: Greenlight already costs $100 to submit a project. Not many people know this that haven’t actually gone through the process. This means that if Valve chooses the low end of the spectrum, it will not be any different from a cost perspective.
Point 2: $100 – $5,000 is easier to come by than 10’s of thousands of votes required to pass the Greenlight system. This means you will inevitably see more games pass through the new system than they would Greenlight.
Point 3: If you have a game people want to play, then you can find $100 – $5,000 or more through many different means including (and not limited to) part-time employment or crowdfunding. This means the cost to enter Steam will most likely not filter out more than the current Greenlight does.
Point 4: The fee is stated to be “recoupable”, which would lead one to believe that Valve will probably have a system in place to not take royalties from the first income from the game until the amount it recouped. This means that developers are not actually losing any money from this fee unless the game sells so poorly it can’t even recoup the fee. If this is the case then the game shouldn’t be on the store anyways.
Point 5: The Discovery Queue and Greenlight Queue are almost identical. The only difference is that in a Discovery Queue you can actually BUY a game, which is much more important to both:
- Players, because you get to play right away, and…
- Valve, because they only make money when you buy a game, not when you vote on Greenlight.
Machine Learning Is The Future!
After glancing through the comments of the original post I noticed most everyone focused on either hating Greenlight because it created “crappy games” or were freaking out over the potential cost of game submission. No one seemed to notice the most important message:
The Discovery Queue will be the main filtration of what games you see and buy!
The Discovery Queue is similar to many other online stores like Amazon and Google. It uses machine learning to figure out what games you’re most likely to buy, and then shows you those similar games so you actually buy them. You help it learn what you want and don’t want by either buying the game, or telling it to not show you it again. This may come as a surprise to some, but Steam is a store that wants you to buy as much as possible!
Valve clearly boasts in this write-up that they are very happy with the success of the Discovery Queue…
One of the clearest metrics is that the average time customers spend playing games on Steam has steadily increased since the first Discovery Update. Over the same time period, the average number of titles purchased on Steam by individual customers has doubled. Both of these data points suggest that we’re achieving our goal of helping users find more games that they enjoy playing.
That number is very impressive, and not an uncommon gain for machine learning algorithms in stores. This is why you get targeted ads on Youtube and similarly bought items on Amazon. It’s an effective system that delivers you more of what you actually want. Target’s algorithm even predicted that a girl was pregnant before she even knew herself by cross analyzing the items she purchased with other pregnant women. Machine learning and targeted advertisements have changed the way we as consumers live our daily lives, and it will be how we as gamers discover our new favorite games.
The Discovery Queue Is Your Game-Introducing Hero
Whether you love it or hate it, Valve’s goal is very clear: Allow as many games as possible into their store and only show you what you like to buy.
Valve believes that by creating a machine learning algorithm, you can decide what a “quality game” is and then get game recommendations based on your interactions with the Discovery Queue. This is the true genius of machine learning algorithms, because every person is a unique individual with different tastes.
We will no longer discover games by scrolling through an endless list, but instead by getting intelligent and curated suggestions. Hopefully this is a positive influence on players as consumers, and on developers to allow them to more easily connect with new fans.
What do you think?
I hope you found this information regarding Steam cancelling Greenlight insightful. I love chatting with fans, so please let me know what you think by following me on Twitter @Disastercake and on Facebook.