It turns out making good videogames is hard even for trillion-dollar tech corporations

A New World screenshot.
(Image credit: Amazon)

Google and Amazon have changed the way the world works, but it doesn't look like they're going to change the way people make good videogames. Some things, perhaps, can't be disrupted.

Google has completely given up on plans to create its own Stadia-exclusive games, citing the increasing cost of development (which doesn't remotely sound like the whole story). Amazon is staying in the game, although after six years of trying it still hasn't released a hit. It keeps delaying its big MMO, New World, and has canceled multiple other games, including Crucible, which bombed last year. Reports from Wired (opens in new tab) and Bloomberg (opens in new tab) tell a story of leadership that supposes that technology, data, and the right management principles will result in billion-dollar franchises—but it hasn't worked yet.

If you're an established game studio, you've got to be feeling at least a little smug that two companies worth over $1 trillion each haven't figured out how to release a single good videogame.

I wonder whether the spirit of creativity is being well-nurtured by having slogans such as "customer obsession" and "deliver results" plastered on a wall.

For years, the accepted wisdom about game development has been: "It's hard." A promising project from a star developer releases in shambles? Well, game development is hard. It would've been awfully embarrassing to have to admit that everyone was just doing it wrong until dot-com billionaires showed up with the right management aphorisms. Instead, Amazon has had a hard time like everyone else. If you're a studio like BioWare that has struggled to release a hit in recent years, that had to come as some relief—not so easy, is it, Bezos?

That's not to say that no one can be faulted: At big studios, the rule about game development being hard is often modified to note that it's especially hard when management sucks. In Amazon's case, releasing a successful game might just require a different management style than building an ecommerce platform, and the "bro culture" reported by Bloomberg (opens in new tab) is discouraging. 

At a basic level, I also hope that making a good videogame requires something that is not technology, or data, or project management strategy at all—something enigmatic and ephemeral, a 'creative energy,' whatever that means. Even in a big business, there has to be a little mystery and playfulness to the creative process, right?

That energy might exist at Amazon. I visited the studio a year ago, and it seemed like a building full of creative people trying to make something cool, like most studios. I do wonder, however, whether the spirit of creativity is being well-nurtured by having slogans such as "customer obsession" and "deliver results" plastered on a wall.

A wall inside Amazon Game Studios' Irvine office. (Image credit: Future)

According to Bloomberg, Amazon Game Studios head Mike Frazzini likes to say that "anything can be measured," which isn't the most inspiring view of game design. In contrast, original Xbox designer Seamus Blackley says that "nobody who's successful starts with metrics." Blackley is clearly winning that rhetorical argument at the moment, but is there truth to Frazzini's point of view? 

Well, maybe a little. Valve is all about data, after all, even biometric data, and its games are always hits (well, except for Artifact). Another seeming point for Frazzini is the success of Riot's Valorant, the pitch for which sounds like it's coming from a stock advisor talking up a recent IPO: It's based on the proven Counter-Strike formula, includes a commando guy, a sneaky guy, and a bow and arrow guy (three of the top videogame guys), prioritizes anti-cheat and networking technology (big gamer concerns), and looks sort of anime-ish (the favored art style among the 18-25 age demographic). It was grown in a lab to be popular.

But I don't think that either Valve or Riot's success can be attributed primarily to market research and user data. Data didn't write GLaDOS's lines, and market research didn't fumble around with ideas for new Half-Life games for years and then get excited about VR, help pioneer the technology, and then make a great VR Half-Life game. As for Valorant, it's not a game that just anyone with a copy of CS:GO and a budget could have made. There's something alchemical to its success, too, a brilliance within Riot that turned 'CS:GO but with characters' into a fun game when it easily could've been an embarrassing flop.


Valorant may not have the most original cast, but more than a copy-and-paste of better games. (Image credit: Riot)

At the risk of endorsing the idea of 'BioWare magic'—a mysterious property of the universe that's supposed to turn struggling BioWare projects into hits at the last minute, which notoriously failed to manifest before Anthem shipped—there's got to be at least some feeling and intuition involved in the search for fun. 

It may turn out that Amazon does possess that creative something. Its struggles just confirm to me that data science and management slogans can't sub in for it. Like every other developer, Amazon is dealing with the incontrovertible fact that making games is hard, and that making successful ones is seemingly harder than ever, given that your audience might just decide that it doesn't give a crap about the $40M you spent on marketing, because it's busy playing a janky co-op game about ghost hunting.

It's like the concept of videogames was so alien to Google that it glitched out on contact.

There are no sure things in game development, and that's true whether your company is valued at $1.7 trillion or you're a tiny developer no one has ever heard of. You can both release a game on Steam—on the same shelf, so to speak—and there's no analytical way to tell which is going to succeed.

Owning Twitch isn't going to hurt Amazon's chances, of course, and being a giant corporation is an advantage in that it can keep on trying forever. It took Activision two tries to turn Call of Duty into a battle royale hit—remember Blackout?—but most companies aren't big enough to get any do-overs. Amazon's next CEO, Andy Jassy, is reportedly committed to trying and trying again to make a hit game. 

"Though we haven’t consistently succeeded yet in [Amazon Game Studios], I believe we will if we hang in there," Jassy wrote in a company email, according to Bloomberg (opens in new tab).

As far as Google goes, who knows what it's thinking? It cannonballed into cloud streaming with Stadia, made a terrible ad (opens in new tab), hired Jade Raymond to make a game, and then chickened out of game development entirely. It's like the concept of videogames was so alien to Google that it glitched out on contact.

Google and Amazon may well have the last laugh. Making games is just a side project for them: They want to control gaming itself. That's why Amazon bought Twitch, and why both companies got into cloud streaming. Maybe they'll succeed in the end, and maybe New World will be the hit that gets things rolling for Amazon. I've played it a couple times, and I think it's the company's best bet yet. As it has for a while, though, it still feels just as likely a company we've never heard of, or one person working out of their bedroom, will make our next favorite game.

Tyler Wilde
Executive Editor

Tyler grew up in Silicon Valley during the rise of personal computers, playing games like Zork and Arkanoid on the early PCs his parents brought home. He was later captivated by Myst, SimCity, Civilization, Command & Conquer, Bushido Blade (yeah, he had Bleem!), and all the shooters they call "boomer shooters" now. In 2006, Tyler wrote his first professional review of a videogame: Super Dragon Ball Z for the PS2. He thought it was OK. In 2011, he joined PC Gamer, and today he's focused on the site's news coverage. His hobbies include amateur boxing and adding to his 1,200-plus hours in Rocket League.