On the one hand, entertainment products are difficult to capture by the current startup rhetoric because (1) the audience/market is broad, (2) the user doesn't have a problem (well, perhaps "boredom"), both of which put a damper on the customer development method.
Nonetheless, there's still a lot to apply. Identifying early adopter players and a channel to them, measuring retention and other analytics, iterating and releasing frequently, creating a viral loop -- these are all concepts that apply to game dev.
At some point, I think most games do try a "big launch" because ads and top 10 lists is how a majority of people find new games (even indie games now have the Humble Bundle as a consolidated channel). Another reason for big launches is the fake market dynamics of the app store: a one time boost when launching, and a positive feedback loop for downloads. But even with big launches, a lot of testing and tweaking is done internally (and it's standard practice on the app store to test the release in single markets, likes say Canada, because the app store stats are independent for each market)
Really good info. I'd love (and even prefer) to release my game today so I can get feedback earlier. But in its current state (shitty graphics, no tutorial, 50% of the intended features implemented) I'm lost as to if that would be good or bad. Common sense dictates the best way is the "big launch" you're talking about.
I think I can make a good/great game eventually given a long enough time with many feedback cycles, but I'm not sure if the user base for games ever grow that way. It seems that you get your one chance to be amazing from the start or you wasted all your time.
My plan is to get to a point where I can have people in coffee shops try it before I make it public or have a very tiny private beta eventually. But I wish there was a way I could get feedback sooner from more than friends and family.
Minecraft was publicly released as an early alpha in 2009, and there was not a full release until 2011. Thisstands as an example proving that a clearly unfinished game can be released and iterated into a much better game than the developer could have created in isolation.
The trick was that the state was described honestly. It was offered cheaply. Upgrades were always free, but the next version always cost more. This set expectations properly.
If Notch can do it, and you want to, maybe you should try it.
