Steamy Refunds (Or, Why We Care About Steam Refunds)


Those readers who use Steam (so… nearly everyone) are likely aware of Steam’s policy on refunds. Namely, that they don’t give them. Steam updated that policy this week, however, and it is… considerably broader.

Pulling a very random (and unprompted) u-turn, Valve has announced that they will now give refunds “for any reason.” There are stipulations to this, of course, but they are not particularly stringent ones: Refunds are automatically given to anyone who requests one within 14 days, and provided the individual has not put more than two hours into a game. If neither of these requirements are fulfilled (i.e. someone has had a game for longer than two weeks), then Valve will evaluate the refund on a case-by-case basis. The page explaining this does not detail how these cases will be evaluated.


Downloadable Content works on a similar rule, provided the DLC is not used up when it is purchased (e.g. extra lives), has not been owned for more than 14 days, and the game it applies to has not had more than two hours put into it since downloading the DLC. Valve cannot guarantee refunds for third-party DLC, rather understandably.

Pre-orders and adding funds to a Steam Wallet can also be refunded – pre-orders can be canceled at any time before release, and can be refunded after release provided it has not been 14 days since release, and that two hours have not been put into the game.  For Steam Wallet refunds, it is the rather simple policy of “If none of it has been spent, you can have it back.”

If any of this has made readers angry, or at the very least confused, don’t worry, we’re with you. For a majority of AAA games, this new refund policy is simply a great consumer tool. For our lovely Indies though, it’s a bit of a slap to the face. Plenty of Indie games are shorter than two hours, and though Valve claims they have an anti-abuse system in place, it will inevitably cause a loss in sales, since even a small percentage of people playing a short game and then getting an automatic refund will cause a noticeable drop in profits to developers for whom every sale matters. Not surprisingly, some devs immediately reacted to the situation, including Black Ice developer Garrett Cooper.

SteamRefund(2)The problem with Valve’s new refund system is not that it makes things better for consumers (because it clearly does, and as a consumer, that’s good). The problem is that their refund policy has gone from zero tolerance to overtolerance, and it represents a serious problem for anyone whose game can be completed in under two hours. Relying on The Honor System is simply something that doesn’t work out when it comes to relationships between businesses and consumers. Plus, Valve isn’t exactly the best at customer service.

Disagree? Think Valve will have enough control to make this work? Any other thoughts? Let us know in the comments.

A nerd of elephantine proportions (both figuratively and literally), Connor also writes for Pxlbyte, and has recently come to realize that he is, in actuality, really bad at video games. So he writes about them instead.

  • 127wexfordroad

    Abuse happens everywhere, so I am sure some people will abuse the system for as long as they can, but generally I think people are going to use it as intended. If you get a game and it is not at all what you feel the game was advertised as in a way that prevents you from enjoying it, then you’re going to return it. Otherwise, you’re going to keep it, because most gamers are in it for the love of the games.

    Indie devs should be no more or less afraid than big studios. I am just as likely to return a AAA title (maybe more, because big studios have the means to make theri every whim come true) as I am an indie title. If a game is good, though, I’m not going to return it.

    So the moral is the same as ever: make a good game, be honest in how you advertise it, and people will buy it and play it. Lie to your customer, or abuse their trust, or don’t maintain your game, and lose the sale.

    I think that’s very fair.

  • The solution is surprisingly easy: Transparency. Which Steam users have requested and received a refund should be publicly shown on the game’s Steam page, including their usage stats prior to the refund. There’s nothing like shame to keep people in line. The true sociopaths, of course, won’t be swayed, but fortunately they’re in the minority of minorities. If that augmentation of the “honor system” isn’t sufficient, then more Draconian measures should be taken. All in all, I think refunds are a great and fundamentally necessary option. I mean, if the game sucks or if you like “trying” a lot of games and were genuinely dissatisfied (as with in nearly all retail stores) a refund is a basic consumer right. I imagine that Valve will quickly adjust its refund policy as necessary according to its own observations and consumer/developer feedback. A little pain to grow properly, hopfully.

  • Agreed… I think of it similarly to DRM-free indie games. Yeah, your little game could show up on TPB with mad seeders and you’d be torn between happiness and anger, but that’s just how things are. A good game just won’t be kept down, and to rail against small slights like abusive refunds or piracy is exactly what the MPAA and AAA publishers do. We’re indie; we are better than that. :)

  • Keeping an easily accessible public record of usage stats and return ratios is an interesting option I hadn’t considered. I wonder how the general public would react to that?

  • I wonder though: Most folks wait until Steam sales or PS+ to get their indie games for as close to nothing – or actually free – as possible. I’m torn between thinking the community won’t bother with refunds if they’re only paying pennies for indie games anyway, and wondering how many will see this as a challenge to complete games in under 48 hours just to “score” a refund.

  • Connor Selinske

    Like Vinny is saying, and like economics tell us, people will always try to get as much as they can for as little as possible. It’s why the Humble Bundles are popular, and also why so many people are worried about refunds. The idea that “Devs just have to make good games” only goes so far. If there’s a good game, but it takes less than two hours to complete, there will be abuse through the new refund system.

    Don’t get me wrong, this is a good thing for consumers, and can potentially be useful for developers of larger games as well, but the tiny experiences may vanish from Steam.

  • Connor Selinske

    I do think you’re right, refunds are extremely necessary for a healthy ecosystem, but I don’t think Steam, particularly with so many Indies and Beta/Early Access programs is a “One-size fits all” kind of deal. The ideal right now would be to see Steam give developers more control over their own refund policies, though probably this would involve allowing an appeal of sorts to Valve, to supercede any devs actually trying to scam consumers (though on the whole, I don’t think this comes up as a problem).

  • Connor Selinske

    I think it would turn out primarily like a review system? It’d be more honest in some ways, but also incredibly useless in others, just like other bits of data. As with everything, there’s no accounting for taste. A taste accountant would probably be paid extremely well though.

  • I wonder if it would actually deter people from gaming the system though. I wonder how many folks care about their reputation on Steam, as opposed to those who generally keep to themselves and don’t participate much within the community.

  • Got an idea… Steam could take the average price paid for a game and divide that by the average time played, producing an “average dollar per hour” metric which it would use to determine the game’s baseline refund value. This could be used to determine how much a person will be refunded (capped at 100% of the individual purchaser’s sale price) based on how long they played the game.

    An example: A game on average takes 10 hours to complete and on average costs $20. The dollar/hour metric would then be $2. If you bought the game for $10, played for two hours, then asked for a refund, your refund would be $16.00 ($20 – twenty percent of $20). You’d get your refund, but not for the time you played relative to the average time played by others users and adjusted for the average price paid for other users. This would prevent people from finishing the game and asking for a refund unless they finished the game significantly faster than the average play time of other users, which unless the game is not fun or reasonable to finish would be difficult.

  • That’s an interesting idea. I particularly like that is forces the gamer community to be responsible for the metrics that determine their own refund. I’d like to see this tested in some capacity.