Gabe Newell on monetization before game design
joystiq.com
joystiq.com
They've started a big push marketing the game, they obviously want people to stay excited for what it is, rather than thinking about any downsides. His answer doesn't mean they haven't thought about it, it means one of two things: either they are yet to come to a decision, or it won't actually be free-to-play. Either way, answering that in this interview would have been a moronic step to take, so he changed the subject away from it, and onto "we care more about games than money".
I love Valve, they're a wonderful company, but they definitely think about monetisation right from the start of a project - they're not idiots. And making money from a game doesn't have to mean the game isn't good.
He said they care about the games first, i.e. ordering over time. I quote, adding emphasis to his time-based terms (four times):
> "and then we'll worry about monetization ... we're not going to worry about that until later." Further pressing the point, Newell adds, "Premature monetization is the root of all evil."
> "I think not sucking is way more of an important thing to pay attention to first,"
As an example of making something cool first, and monetizing later, consider google. They didn't do ads for several years, until they found a way to do it that didn't conflict with what made their search cool.
BTW: kevingadd's perceptive, reasoned and evidence-based reply to you is [dead] - this is disturbing to see on HN.
I don't think that's actually true. If you look at a lot of their past projects, the real clever monetization comes after release, and very often they release piles of cool content entirely for free when they could easily have charged for it. They also spend much longer in development than is strictly necessary, and that tends to not make sense if your goal is to earn money efficiently. As a company they clearly understand monetization, but I don't think it drives their development process.
I just don't see it.
Like "build something people want".
One way to looks at it is in terms of dependencies. Complicated problems have a ton of dependencies, so that when you change one thing, it ripples through, changing other things. Therefore you'd best get the most important things right first, leaving everything else loose, flexible and uncommitted, so that as you fiddle with the important things, you aren't tied down by other issues. I find I can easily become overly fond of some secondary aspect that prevents me backing out of that local minima and seeing the true (or truer) truth. Once you've got the key things right, you can start worrying about secondary things (secondary in that they are still important, just not as important). So it's similar to the knuth quote alluded to (http://c2.com/cgi/wiki?PrematureOptimization). (though I'd say that the most important thing is understanding the problem and solution, rather than having correct code).
It's really good to just completely forget about making money, and set out purely to make the world a better place (or realize a vision). It's freeing.
"I don't know about other game companies, but it's dangerous to kvetch at Valve. You're suddenly Director of Fixing That Shit! Vice President of It's Your Problem Now."
I don't look at it as money vs quality. I see it more as, more quality more money. So start with quality.
The only way to make money is if people play your games. The perceived value of your games is higher, and the amount of interest is higher when you make good games.
One thing Valve has, probably more than any other game company (besides possibly Zynga) is a ton of data on gamer behavior across all their titles.
They've looked really hard at how players interact on some of their AAA titles like Team Fortress 2 and Left 4 Dead, and more recently on Portal 2. They know how people play, and what they're willing to spend on.
Seriously, they are running EEG brainwave scans and pupil dilation tests on gamers during play testing to understand how their games are effecting their players. That level of customer understanding makes it pretty easy to know where and how to turn the money machine on.
The product is the only selling point in Valve design.
Even before Valve had ever shipped a product, this is how they worked, and you could argue that their customers rewarded that approach.
Perhaps the simplest fix is to just make the same comment again.
No one's suggesting that you shouldn't think about monetization, he hasn't even launched his product yet.
Now what would be stupid is doing like some founders and giving away the product gratis for a year or two and then suddenly demanding money. That would be stupid.
Getting something worth selling before worrying about what you'll charge for it isn't.
But I bet the millions of dollars from working at microsoft helped a bit too in giving him a long perspective.