Kevin Geisler, a developer over at Young Horses, has just written up a post over on Gamasutra reporting that 30% of players have used Valve’s new Steam refund policy on Octodad: Dadliest Catch since the policy was introduced.
The Steam refunds system was put in place on June 2 and has seen a lot of warring opinions on some of its specific policies. It allows Steam users to return any game for any reason within two weeks as long as there has been less than two hours of playtime. Many feel that the time limit is arbitrary and unfairly punishes shorter games where two hours could be a significant chunk of the total playtime or even the entire experience. Octodad is just such a game, so a 30 percent return rate initially seems like the exact scenario people were afraid of, and you might expect it to be hitting the developer’s pockets pretty hard—not as bad as you might expect, says Geisler:
Despite total refunds being 30% of units, it ends up only accounting for a loss of ~13% net revenue when looking solely at this week. These numbers sound high, but consider that we’ve sold a lot of units in the last 6 months and our current week is at our typical ‘tail’ rate (much lower) when the game is not on discount.
Geisler’s posts highlights another interesting fact about the refunds of Octodad:
Since Steam does break down the data by sale price, we are in a position where we can have some certainty on when a returned game was purchased. For instance, we participated in the winter sale in December at 50% & 75% off, and in a ‘Midweek Madness’ sale in April at 66% off. It turns out that 80% of our refunds match the sale price for these periods, putting the purchase dates between 2-5 months ago. [Emphasis his.] The other 20% of refunds at full price could have been purchased at any point in the last 6 months.
The two-week time limit doesn’t seem like it’s getting enforced at the moment and it’s hard to say why just yet. Valve could be simply allowing customers to return games purchased months ago temporarily since the policy was only just implemented. As well, Geisler points to a reddit post showing an anecdotal example of a player able to get a refund for a game he played for six hours—far exceeding the policy’s two-hour time limit. At the same time, Valve’s policy is strictly on “a case-by-case basis,” so it probably shouldn’t be too surprising that exceptions are being made.
It’s an interesting post that brings real data to what has otherwise been a fairly hypothetical argument. Geisler sounds confident that if players are found to be abusing the new system, either those players will be stopped or the policy will be changed, and that’s probably a good way of looking at it. Valve’s hands-off approach hasn’t necessarily yielded the best results in the past (see: the flaming mess that is Steam Greenlight) but that they’re tackling the long-overdue issue of refunds at all is fantastic.