Google's employment roster....

Teamwork is the norm, especially for big projects. Keith Coleman, a 26-year-old product manager who works on Gmail, oversees a ten-person secret project whose team embers have taken over heir own conference room. They've given up their big space to be crammed into this room to get things done," says Coleman. The hideaway happens to be where Gmail's chat function was created. Lounge music is usually playing, engineers wander in and out, and there's no formal daily meeting, though the team tends to congregate between five and seven in the evening. "If I could capture anything that's great about Google," says Coleman, "it's that room."

Elsewhere on campus, Google has been methodically picking off experts in whatever field it desires and adding them to its collection of employees. Much has been made, for example, of the fact that Google now employs Vint Cerf, the "father" of the Internet, and Larry Brilliant, who helped eradicate smallpox and now runs Google's fledgling philanthropic arm. But the bench strength runs even deeper, including nonemployee advisors. One day Coleman, the Gmail product manager, noticed his former professor, Terry Winograd, a renowned Stanford researcher of human-computer-interaction design, roaming the hallways. Now Winograd, a Google consultant when he's not teaching at Stanford, sits in on Coleman's product bull sessions.

When they join Google, all the engineers receive a copy of Essential Java, written by former Sun engineer Joshua Bloch (the above-mentioned fan of roast quail). He joined Google in 2004 as chief Java architect. Engineers at Google swear by Python, a software-writing language. So last year the company hired Guido van Rossum, Python's author. Van Rossum brags on his personal Web page, "I knew Google's services to be awesome; to work there is even better. But best is that I get to spend half my time on Python, no strings attached!"

Van Rossum is a special case, but all Google engineers are famously required to devote 20% of their time to pursuing projects they dream up that will help the company. The projects actually have a realistic chance of being adopted too. Google News, Gmail, and the Google Finance site all sprouted from 20% time. Nontech ideas Googlers dream up have a shot at adoption as well. The Google shuttle bus exists today because Cari Spivack, who used to work on the company's book-search product, got sick of driving to work, scouted out a bus company, then plotted out the routes a shuttle might take. She brought the idea to senior management only after she'd done all the research. Any other company Google's size would form a cross-divisional transportation feasibility committee to study the issue. Google just did it.

That's not to say that anything goes at Google, where even the smallest issues are open to academic-style debate. On pajama day, in the all-organic No Name Café, I spot Sheryl Sandberg, the company's all-business vice president for online sales operations. Like perhaps a tenth of the employees I see, she looks as if she has just rolled out of bed (she's in turquoise flannel PJs with bears riding space rockets). But nearby is a coterie of engineers wearing tuxedos: In the spirit of debate, they're protesting pajama day.

Google has a name for its gazillionaires: "economic volunteers." The term refers to the several hundred employees who received gonzo option grants in the company's pre-IPO days. Just as Microsoft did for years, Google initially paid below-market salaries because everyone believed, correctly, that the stock's upside was so large. Lately, Google has begun paying salaries that are competitive with the rest of the tech industry. Experienced engineers can expect to make as much as $130,000 a year and receive 800 options and 400 shares of restricted stock when they join. New MBAs can expect between $80,000 and $120,000 and typically receive smaller option grants.

Though Google (Charts) almost always wins talent wars with the likes of Microsoft (Charts) and Yahoo (Charts), its twin HR demons are startups and retirement. Take the case of Niniane Wang, a 27-year-old engineer who received a million-dollar founders' award for her work on a program that searches computer desktops. She says she loves Google, puts all headhunter calls on "auto-reject," and isn't thinking about leaving ... but if she did leave, it would be to start her own company.

Wayne Rosing, an early vice president for engineering, has left to pursue his love of astronomy. Paul Buchheit, the celebrated engineer who dreamed up Gmail, "retired" recently. (He's 30.) Evan Williams, the highly regarded founder of Blogger, a Google acquisition, has left to launch another startup.

To stave off the inevitable trickle for the exits, Google is bringing in HR pros like Laszlo Bock, an ex--GE executive who joined the company earlier this year as vice president for people operations. Though the current corporate attrition rate is around 5% - and a more closely watched internal measure of employees whom Google truly regrets losing hovers around 22% - Bock's group is particularly concerned about mass vesting. At the end of 2002, Google had nearly 700 employees. All those employees - if they're still around - are now fully vested on their initial stock-option grants. Many of them received a generous batch of stock options in the spring of 2003, which will fully vest this spring, making them a flight risk. Already there are people who are coasting: Around campus they're said to be "resting and vesting."

The solution? Google is considering starting a sabbatical program and thinking of new career opportunities within the company for the restless. It has also institutionalized a bevy of compensation incentives, including restricted stock that immediately has value to new employees, founders' awards that can run into millions of dollars, and special bonuses. And it is about to unveil a new facet to its options program: Going forward, employees who've been around long enough to vest will be able to sell out-of-the-money options on the open market. Though no other company has tried such a system, Google says it is confident the Securities and Exchange Commission won't have an issue with it.

Another danger is harder to address: growth. It's easy to feel like an outlaw band that's changing the world when you have 100 employees. It's incredibly difficult when you have 10,000 or 100,000. Many Silicon Valley types, after all, don't like working for big companies because geeking out and trying new things becomes a hassle. Sergey Brin identifies with the concern. "Before, if I wanted to change a few lines of code and push it to the site, I'd just do it," he says. "The users would be the testers. We can't do that today."

In that respect, Google's in the same boat as any big tech company. But unlike, say, Microsoft or eBay, it does have some protection from the demands of the outside world. In a move seen as controversial when they unveiled it before the IPO in 2004, the founders retained voting control of the company, meaning that as long as Brin and Page are in charge, they get to decide whether the upkeep on the lap pool and climbing wall is worth the expense.

The one thing the founders can't control is how long Google stays on top. The Valley is littered with former "it" companies that slid into obscurity or devolved into middle-aged mediocrity. Which is why Google's founders have sought out a role model for building their culture, and it's not a tech company or an ad giant. It's Genentech (Charts), the biotech company that is No. 2 on our Best Companies list (and was No. 1 last year). Genentech has seen drugs succeed wildly and fail miserably, has had years when its stock soared and years when it sank. But through all its ups and downs over 30 years, the company has remained a scientist's paradise and a place where people love to work.

The year Google went public, CEO Eric Schmidt welcomed Genentech's CEO, Art Levinson, to Google's board. Levinson likens his introduction to Google's management troika to his first meeting, in 1979, with Genentech's founders, who promised him he'd "save the world" by joining Genentech. "It's very similar to the message I heard from Larry, Sergey, and Eric," he says. "They see themselves as poised to do something radically different from what has ever been done anywhere else." In other words (except for that curing-cancer part), just like Genentech.

So back to that $483 question: Is Google's culture great because its stock is doing well, or is its stock doing well because its culture is great? The answer, of course, is that you can't answer the question when Google's stock is at $483. Like Genentech or any other company that's been through the wringer without losing the loyalty of its staff, you can gauge the impact the stock price has on the culture only once the stock takes a dive - or stalls for an uncomfortably long time (see Microsoft, 2001--06).

There are, however, encouraging signs, says Jeffrey Pfeffer, an organizational- behavior professor at Stanford. Pfeffer recalls lunching with Page several years ago, and even at that point the cofounder was serious about creating a corporate culture at Google. "They have been extremely thoughtful about this," he says. Pfeffer, who has done research on the money that companies save in the form of productivity gained by providing benefits that let employees focus on work, says Google's perks fit his research thesis to a tee.

Art Levinson, for his part, sees a parallel to Genentech that goes far beyond any perks. "What draws people to both companies is the environment, one where they have an ability to pursue things largely on their own terms," he says. Of course, there is one more reason Levinson is a Google acolyte: "Here I am, a guy who can afford a good meal, and every time I go to a Google board meeting, I don't leave until ten o'clock at night because I get a free dinner there."

At Google it really does always come back to the food.

______________________

Best Companies to Work For: See the full list 

1 2