And while Bell Labs and others made huge contributions to basic, university-style research, X projects are conceived as moneymaking enterprises, or things that at least seem as if they could make money sometime in the next few years. With Foghorn, the goal was to turn seawater into gasoline for no more than $ 5 a gallon â enough that it could conceivably find a market in some European nations where high taxes make gasoline more costly.
Kathy Cooper, an engineer who led the project, said of the goal, âI think we could meet it, but it would be more like 15 to 20 years.â That was too far away, which was why she recommended killing it.
X employees avoid talking about money, but it is not a subject they can ignore. They face financial barriers that can shut down a project if it does not pan out as quickly as planned. And they have to meet various milestones before they can hire more people for their teams.
The point, several executives said, is to encourage employees to think big while staying away from creating technology for technologyâs sake.
âIf you actually want to make the world better, focus on what would actually make the world better, and the technology will take care of itself,â Mr. Teller said. In other words, great ideas that solve real problems will bring in real money.
Credit Jason Henry for The New York Times
One of the more public examples was a recent overhaul of a Google robotics effort whose best-known project is a robotic cheetah. The division was recently dismantled, and its disparate technologies are being re-evaluated so they can either be sold off or streamlined for a specific moonshot idea, like âbuild a car that can drive itself.â
Over the last year, X has hardened project milestones and created an intermediate step, called Foundry. The groupâs leader, Obi Felten, said her job was to hone new technologies into a concrete product and to work with engineers like Ms. Cooper to help them write business plans.
The idea is to create a conveyor-belt-like process that encourages employees to quickly jettison projects that seem unlikely to work and to avoid technologies that are neat ideas but would never make money.
One of the best ways to save money, Mr. Teller said, is to encourage employees to kill projects before they become expensive â hence the bonus for the Foghorn team. X staff meetings begin with a âpremortemâ process in which people predict how various technologies might fall short. Employeesâ laptops are decorated with stickers of crumpled-up paper that represent the end of past efforts.
When Mr. Teller speaks at events like South by Southwest and TED, he ruminates so deeply on the word âfailâ that successes are sometimes framed as âfailing to fail.â
Video by TED
The idea of celebrating failure is a Silicon Valley clichÃ©, but Mr. Teller talks about it in the practical terms of a management consultant. Say you have a team of 20 people working on a project that is not going anywhere, he said in a recent interview. In a year those 20 people will be 30 people. The company has to pay their salaries and health insurance, and the team will inevitably hire a few consultants. Worse, they will have wasted a year.
How much money could the company save if you could get them to cut bait a year earlier?
âThatâs the magic,â Mr. Teller said. âBut they wonât do it. They will not raise their hands and say, âThis project is just not what we should be working on,â unless you start bonusing them, unless they can get a promotion for ending their project.â
Failure bonuses are also an example of how X, which was set up independent of Google from the outset, is a leading indicator of sorts for how the autonomous Alphabet could work. In Alphabet, employees who do not work for Mother Google are supposed to have their financial futures tied to their own company instead of Googleâs search ads. At X, that means killing things before they become too expensive.
Credit Jason Henry for The New York Times
Mr. Tellerâs decision to start paying bonuses to employees who kill their own projects happened long before the Alphabet transition, and it was met with skepticism from Googleâs chief culture officer. Mr. Teller got what he wanted anyway.
âThey were just offended by the idea,â he said. âAnd now we donât even talk about it anymore.â
Not a Bunch of Techies
X was founded in 2010, and it was originally in a brick building on the edge of Googleâs main campus. At the beginning, it felt like an extended 20 percent project â the extra projects Google employees are encouraged to work on outside of their main job â with little in the way of process.
Former Google employees said this led to a temptation to overhype projects long before they were ready. The most notorious example of this was Google Glass, the much maligned computer glasses.
The glasses were supposed to have a marketing scheme to match, and X employees hatched an idea for floating showrooms. Later, when mysterious barges showed up in Portland, Me., and San Francisco Bay, they kicked off months of speculation and conspiracy theories until finally they were towed away without a word from Google.
Mr. Teller said he shut down the barges idea the first day it was his decision to make. âThere wasnât a lot of agonizing about that,â he said.
Today, the companyâs offices are a few miles from Googleâs main campus. They are full of electric vehicles and people riding skateboardlike contraptions. Mr. Teller appears to be always on Rollerblades.
Beyond the main lobby and offices, behind key-card-protected doors, sit various labs, as well as an open warehouse floor that houses a division called the Design Kitchen, which appears to be a cross between a machine shop and a cool dadâs garage. Some items spotted on a recent tour included climbing rope; lots of tools; oversize plugs; a set of skateboard wheels attached to a block, wires and a 9-volt battery; a plastic ear; and beer.
Credit Jason Henry for The New York Times
Unlike Google, where software engineers are kings, X is run by an eclectic batch of scientists and tinkerers whose backgrounds range from math, physics and chemistry to design, fashion and public art. The line âIâm not an engineerâ seems to be a point of pride. Rich DeVaul, who heads a ârapid evaluationâ team whose job is to hatch and kill ideas in short order, guessed that fewer than half the employees have engineering degrees (he doesnât).
âI would argue that the great strength of X, in fact, is that basically we arenât just a bunch of techies,â he said.
The combination of big ideas, lofty rhetoric and a strict code of secrecy has made X a source of endless speculation and conspiracy theories. The one you hear most frequently, usually from competitors and venture capitalists, is that X is a giant public relations plan to distract regulators from Googleâs search business, which is under scrutiny around the world.
Cynical though that sounds, it points to something that seems fundamentally true: Many of historyâs great corporate research efforts, like Bell Labs and Xerox PARC, have come from companies that were monopolies or close to it.
Waiting for a Moonshot
Failure bonuses notwithstanding, no normal person would walk away from X thinking it looked frugal. There is a free cafeteria whose recent menu included sushi and pork tri-tip. The bathrooms have Japanese toilet seats with buttons for warmth and spritz. Listening to X executives talk about saving money is a bit like having billionaires tell you they built a $ 30 million house for $ 30 million and not a penny more.
But Xâs success will be less about money saved than the creation of a new Google-size business whose profits wash away all the losses. And while the division has already had some successes, it has yet to produce the âmoonshotâ Mr. Teller is hoping for.
Arguably the biggest success, Google Brain, is the one people know the least about. It was an artificial intelligence effort that was spun back into core Google in 2012 and is now embedded in products like voice search and the Google Photos app, in which users can search their albums for images like âdogsâ or âmountains.â
Investors would prefer a shiny new business that does not rely on advertising. Mr. DeVaul hopes that if X can deliver a big hit, it will encourage other companies to copy the entire concept. Management science is a lot less interesting than moonshots, but until X hits the moon, that is what it is selling.
âWe could have five or 10 other world-class organizations doing this, and we would not be able to scratch the surface of the crazy big hard challenges out there,â he said. âIâm an optimist: I believe that there is an endless supply of huge problems affecting humanity.â