Over half of developers surveyed do not think Steam justifies its 30% share of revenues according to GDC’s State of the Game Industry report. Over half of developers responding make 75-100% of their sales on Steam. The difference is not surprising since, just as gamers will think $5 is a fair price for this year’s AAA spectacular, game developers will want to see the price of the service they are consuming go as low as possible. Consumer psychology aside, it is still striking to see these numbers beside each other. The 70/30 split has been a standard for a long time and has recently been called into question with the entry of several high profile storefronts competing on price and store exclusives. Now is a good time to consider what exactly that 30% buys.
A firm wanting to do their own digital distribution must arrange for payment processing, hosting, returns, chargebacks, and then worry about people finding the product. For smaller firms these costs will almost certainly be more than 30% of revenues and so using a digital storefront like Steam represents a savings. In 2019 the choice is not Steam or a private storefront and the question in GDC’s survey was posed in light of competing alternatives that offer a lower price. The profits for digital storefronts have always come from the surplus from the actual cost of providing the services, and the argument for competition is that profit seeking firms will be willing to undercut each other in order to capture more of the market. Assuming that Discord and Epic don’t operate at a loss (not a good assumption in these cases) and that their rates are sustainable, then it is tempting to think that Valve is claiming at least 18% of revenues as a tax for being on Steam. Given that 18% is more than what the new storefronts are charging overall, it makes sense that only 6% of developers surveyed think Valve is justified in claiming that surplus. However, this is not the right way to think about whatever surplus Valve or any other provider is claiming.
People who are good at something often do not find it very difficult to do. The price of someone’s labour is commensurate with the value they provide, not the effort they exert in performing the task. If digital distribution is just a matter of setting up a tool booth and matching developers to gamers then it invites the question why payment processors aren’t opening their own storefronts since they will always be able to undercut the competition by whatever surplus they charge for their own services. For better or worse, Steam is very good at doing this match, which is why almost ¾ of respondents make at least half of their sales on it. In addition to plumbing like payment processing, Steam offers services like the Workshop, Digital Rights Management (DRM), and, even after the increase in games released under Steam Direct, a degree of legitimacy that is not provided by other storefronts. Most importantly, it provides access to a potential audience, even after accounting for complaints about its discovery algorithm. The more favourable revenue shares by competitors do not mean anything if developers cannot get on the storefront or cannot achieve a certain number of sales.
The right question isn’t “Is Valve justifying its 30% cut?” it’s “How do Humble and GOG justify theirs?” Humble has an easier time of this since they essentially piggyback off of Steam for most users (it is also unclear if Humble absorbs whatever costs Steam might charge for external keys). GOG has its own requirements (particularly regarding DRM) which require effort, in addition to stricter controls on who gets in. The fact that only certain games choose to launch on GOG, or appear at all, suggests that at least some developers do not find the additional sales worth the effort to participate on that storefront.
It’s hard to put an exact number as to the right price that any storefront should charge for their services, but at least in the case of Steam we know it’s access to the market that forms the foundation for most of the survey respondents’ businesses and so it is hard to see that number as too high. It may not be fair, but more people are coming to Steam, not less. None of this is to say that discovery is solved or that Steam is the best option for people, but it does make certain decisions by Steam make a lot more sense. Smaller developers were understandably frustrated by the choice to give a break on the revenue share for larger games that pass certain sales thresholds. However, these games clearly do not benefit from the access Steam provides in the same way that smaller titles do, and in the absence of other features to entice these developers, Valve needed to cut its rates in order to retain them. Valve seems to be quite good at matching their prices to developers’ willingness to pay.
This is why discoverability is a big problem for Steam and one they’ll need to continue to work on. If the other storefronts can demonstrate they can connect developers and gamers better than Steam does, then Steam has no reason to charge what it does and will have to decrease their price. Epic seems best positioned to do this as they are attracting an audience that is different from Steam’s and have the financial capital to back it up. And yet behind Epic’s aggressive launch price is an interesting revelation. In the disputed translation of an interview, Epic’s Sergei Galyonkin (author of SteamSpy) apparently revealed that their solution to discovery was to rely on influencers over algorithms and that the referral rate for influencers is set by developers, with Epic covering the first 5% (for now), and an example of indies offering perhaps 20% to incentivize influencers to play the game. Galyonkin is a sharp guy and it is unlikely he chose 20% at random. From this example, large games that require no incentive will face an effective rate of 12% on the Epic Store. Bigger games that need some incentive will face an effective rate of 17% (from purchases from influencers). Smaller games will face an effective rate of 32% from purchases from influencers. Compared to Steam’s tiers of 30%/25%/20%, the Epic store doesn’t seem all that different. In Galyonkin’s example, the value of influencers is about equal to what Valve is charging to automate the process through its algorithms.
Discoverability does not go away once a developer switches storefronts. Valve is using technology to connect its users with games, while Epic intends to replace what sites like Keymailer do. It’s not obvious which approach will be more successful. If Epic’s effective rates are closer to 12% then it means influencers aren’t all that influential and if sales can be maintained to levels comparable to Steam’s then it really does mean that Valve has been charging for a worthless matchmaking service. But if Epic is relying on “ya boy yoloswag420” to connect the game to the audience, it is the developer who will be paying for it, not the creator of Fortnite. If influencers fail to attract buyers to new titles, then developers aren’t going to receive help from Epic in selling their game. In this scenario developers will need to rely on the storefront remaining relatively closed or the 12% cut reflects that the Epic store is just Steam without the algorithms. The competition between Epic and Steam will likely be less about price and more about who has the more effective discovery mechanism.