Google Couldn't Kill 20 Percent Time Even if It Wanted To

Google is reportedly cracking down on 20 percent time. But the concept of employee goof-off time preceded the company and lives beyond it, including at Facebook and LinkedIn.
Image may contain Human Person Shorts Clothing Apparel Footwear Shoe Grass Plant and Restaurant
Photo: Ariel Zambelich/WiredPhoto: Ariel Zambelich/WIRED

Google's "20 percent time" is dead -- or so we worried.

The business website Quartz reported on Friday that Google had effectively killed its famously egalitarian company policy -- which encouraged employees to spend about 20 percent of their time experimenting with their own ideas -- and Silicon Valley has been buzzing ever since. On the website Hacker News, members of the startup set have weighed in on the alleged death some 475 times.

It makes sense that programmers and other knowledge workers would react emotionally to reports of 20 percent time’s demise. Google has become today’s most powerful symbol of a more casual, collegial, and unorthodox approach to management that has been emerging at Valley companies for several decades. Outside the tech bubble, interest was strong enough to fuel production of a major motion picture, The Internship. Inside, the stakes are far higher.

“I'm incredibly sad, as it seems that too many companies go down this road,” wrote one longtime Hacker News denizen.

The only trouble is that 20 percent time is still very much alive.

In part, that’s because Google says its employees continue to create what are commonly referred to as 20 percent projects, despite the disincentives to participation listed by Quartz. We’ll see how that plays out, but in the meantime, there’s a bigger reason 20 percent time will live: Google couldn’t kill the thing if it wanted to.

>There’s a bigger reason 20 percent time will live: Google couldn’t kill the thing if it wanted to.

Google didn’t invent the idea of giving employees time to experiment with their own ideas, nor will it have the final word on how best to bestow such time. Plenty of other large tech companies have implemented their own takes on 20 percent time, including widely admired, innovative companies like Facebook, LinkedIn, and, reportedly, Apple.

The core idea behind 20 percent time -- that knowledge workers are most valuable when granted protected space in which to tinker -- is more alive in Silicon Valley today than it ever has been before, reports of Google killing its program notwithstanding.

The flourishing of 20-percent-style programs shows that, despite a seemingly endless boom-and-bust cycle, with corresponding hiring and firing binges, the technology sector has also developed a lasting corporate culture, one that may yet prove as viral as its products.

I’m not a neutral observer of this phenomenon, having recently written a book on 20 percent time, similar programs, and the recent flowering of side projects in Silicon Valley and beyond. But my interest here is to set people straight on a widely misunderstood topic, not to siphon more money from the publishing industry’s gushing fountain of profits.

To understand why it’s tricky to talk about 20 percent time being dead or alive, you have to first understand what the program is and what it is not. Google’s 20 percent time has historically worked like this: The company allowed employees to spend approximately one-fifth of their time -- one day per week, four days per month, or maybe even a couple of months per year -- working on a Google-related passion project of their own choosing or of their own creation.

This is a Google policy, and it has been mentioned in official documents like press releases, company blog posts, and Larry Page’s and Sergey Brin’s 2004 “founders letter” to prospective Google shareholders. The policy led to products like Google News; Google’s autocomplete system, originally called Google Suggest; Gmail; and AdSense, the advertising engine developed to support Gmail financially, now producing roughly a quarter of Google’s revenue.

What 20 Percent Time Is Not

That’s what 20 percent time is. Here is what it is not: A fully fleshed corporate program with its own written policy, detailed guidelines, and manager. No one gets a “20 percent time” packet at orientation, or pushed into distracting themselves with a side project. Twenty percent time has always operated on a somewhat ad hoc basis, providing an outlet for the company’s brightest, most restless, and most persistent employees -- for people determined to see an idea through to completion, come hell or high water.

For example, engineer Paul Buchheit worked on Gmail for two and half years before he finally persuaded company brass, who worried about stretching Google too far beyond search, to launch the thing. At the time, one member of Google’s founding team even said in a meeting that “this will destroy our brand; this will crush our company,” ex-Googler Chris Wetherell told me for my book.

Wetherell’s own side project, Google Reader, only got placed on Google’s internal servers because Wetherell was friendly with people with administrative access to those machines who could bypass the normal chain of command. (The external public release, which came later, was sanctioned).

>Because 20 percent time is less of a formal program than an idea or operating spirit available to bullheaded employees, availing oneself of 20 percent time has long entailed sacrifices.

Because 20 percent time is less of a formal program than an idea or operating spirit available to bullheaded employees, availing oneself of 20 percent time has long entailed sacrifices. And at a company where bonuses make up a large percentage of income, these sacrifices can be financial, particularly if your manager and co-workers are, for whatever reason, unsupportive of your 20 percent time project. Even if said people could not block your 20 percent time project, they could make you pay a steep price for continuing to pursue it.

So Google has levers it can pull to make 20 percent projects more or less attractive to employees when its enthusiasm for side projects waxes and wanes as it has over the years. While I was writing my book, Google shut down its Labs experimentation platform, and there was speculation 20 percent time would die. When Google VP Bradley Horowitz announced a few months later that “we’re throwing fewer things at the wall,” the company was again asked if 20 percent time was dying, and again said it was not.

But Google has copped in the past to making it harder to ship 20 percent time projects, and Quartz says another crackdown is ongoing. Google says through a spokesperson that employees are still encouraged to pursue what are often called 20 percent projects that involve Google-related interests outside their main line of work. It also says the idea of Googlers running with their own ideas is core to the company’s culture. But it won’t, when pressed repeatedly, state simply and clearly that 20 percent time lives on. So we’ll see.

An Idea Bigger Than Google

In the meantime, remember that the core idea behind 20 percent time didn’t begin at Google -- 3M had 15 percent time -- and won’t die there, because it clearly lives on at other tech companies. One key way tech employees are encouraged to pursue their own experiments these days is at hackathons, programming marathons followed by a presentation of brief demonstrations and, often, prizes. Pioneered by an Australian software company called Atlassian, which was expressly trying to find a cheaper version of 20 percent time, corporate hackathons are now commonplace, including at Twitter, Facebook, Google, Yahoo, LinkedIn, and eBay, to name just a few practitioners.

>Remember that the core idea behind 20 percent time didn’t begin at Google and won’t die there, because it clearly lives on at other tech companies.

Hackathon projects can even graduate to full-fledged products at Facebook, Atlassian, and LinkedIn, all of which offer some sort of pipeline for the most promising hacks. Under LinkedIn’s InCubator, creators of top experiments can get up to three months -- 25 percent of the year if you’re keeping score at home -- to develop their ideas, along with coaching from founder Reid Hoffman.

At Facebook, hacks that make it through a gauntlet known as the “Prototype Forum” and become experimental products like Facebook’s free café Wi-Fi system. And at Atlassian, managers can hand out “get out of work free days” to people on other teams, which, given the right approvals, can snowball into 20 percent time.

Last year, Apple reportedly launched an initiative called “Blue Sky” that allowed small teams to “spend a few weeks on a pet engineering product,” as The Wall Street Journal put it.

None of these programs is as expansive as the public perception of 20 percent time as a universal goof-around-for-free day bestowed on pampered Google employees every week. But then Google’s program was never as expansive as believed. And adding some judging, feedback, and controls to Google’s system, as Facebook et al have done, is very much in the spirit of 20 percent time. What’s the point of an experiment if you’re not going to act based on what you learn?