Inside Google: A Nerd Utopia That Works
from
May/June 2002
by Bonnie Powell
As one of the best-known brands on the Internet, Google could easily have gone public by now. It has significant earnings from diverse revenue sources, an irreproachable management team, and, most important, the strongest search-engine technology available.
The fact that it remains private reveals the conservative heart of this legendarily zany outfit. Despite the lava lamps, scooters, and colorful exercise balls that dot the halls of the “Googleplex” in Mountain View, California, the company has no intention of being tarred with Wall Street’s dot-com brush. Its flirtation with the idea of an IPO last year may have quickened the pulses of bored investment bankers everywhere, but Google seems content to wait until the markets fully recover—or perhaps even longer.
“The IPO question we’ve debated internally, but frankly, we’re profitable. We’re generating cash. We don’t ever need to go public,” chairman and CEO Eric Schmidt, 47, told the San Francisco Chronicle last fall. “The market is sort of dead now, so it’s not a near-term project.” Adds Safa Rashtchy, a U.S. Bancorp Piper Jaffray analyst who covers the search-engine industry: “Google has a totally different mentality from all the other companies here in the Valley. By far, it is more a technological company than an entrepreneurial one. Entrepreneurial companies want to have an idea, take it to market, and then go public. But Google is more obsessed with being the best at what it does.”
By every standard, Google is the most popular search engine on the Internet. It answers more than 150 million queries every day. And amazingly, it has achieved that volume with zero marketing or self-advertising, although the 47 industry awards showered on the company since its 1996 beginning haven’t hurt. Torrents of laudatory journalistic ink have been good for business, too; reporters are big search-engine junkies. [So are editors. —Ed.] The company’s frugality has enabled it to hang on to most of its estimated $36 million in funding and make money from its two current revenue streams, selling keyword-based ads that appear on its website’s results pages and providing search services for sites operated by other companies.
Google’s word-of-mouse popularity derives from the search engine’s uncanny ability to zoom through the Web and produce what you’re looking for in a gnat’s blink. Its index of two billion Web pages is far bigger than any of its competitors’. Fast Search and Transfer (FAST), its closest rival, boasts only 625 million pages. And Google’s two billion doesn’t include its 700 million archived Usenet bulletin-board messages, 330 million images, and countless documents in PDF, Word, and other formats.
To understand Google as a technological rather than an entrepreneurial company, it helps to revisit its birth. Founders Sergey Brin and Larry Page, both still cherubic-looking late-twenty-somethings, are far from typical Internet wunderkinder. While freshly minted M.B.A.’s were scribbling business plans on cocktail napkins, the two Stanford graduate students were concocting papers with titles like “Dynamic Itemset Counting and Implication Rules for Market Basket Data” (Brin et al. 1997). They had Google’s precursor, a search engine called BackRub, up and running in their dorm rooms by early 1996.
Company lore has it that Brin and Page—who never did collect Ph.D.’s—turned Google into a business only because no one was interested in buying what they had built. The initial funding was a $100,000 check from Sun Microsystems founder Andy Bechtolsheim. Friends and family brought the total to $1 million. Already answering 10,000 queries a day, Google was incorporated in August 1998. In mid-1999, Kleiner Perkins Caufield & Byers and Sequoia Capital, perhaps Silicon Valley’s top two venture capitalists, put in $25 million. Kleiner Perkins partner John Doerr and Sequoia partner Michael Moritz joined Ram Shriram, a former executive at Amazon.com and Netscape, as the only outside board members. All three are still on the board. Google, whose corporate headquarters at that point had progressed from the dorms to an actual garage in Menlo Park to offices in Palo Alto, moved into its current Mountain View home soon thereafter.
Meanwhile, its competitors were happily taking the IPO plunge. In July 1999, Ask Jeeves, the natural-language search engine named for the fictional valet and saddled with 1998 losses of $4.3 million, pulled off a $48.3 million offering. So intense was investors’ anticipation that Jeeves stock, priced at $14, actually opened at $70. Says venture capitalist Geoff Yang, a member of the Ask Jeeves board: “The public was saying, ‘Run for the roses—build the brand, build market share, and improve reach.’”
Brin and Page took a different approach. They focused relentlessly on refining and extending their technology. Among other things, Google added wireless search capability, developed interfaces in 66 languages, launched a workable automatic-translation service for non-English search results, and came up with one of its coolest features, a spell-checker that returns proper listings even to the dyslexic.
It also hired as many engineers as the Googleplex could hold. More than 50 of its 270 employees have doctorates, the company says. The bulk of its venture capital has been spent on keeping those employees happy and productive, and on the 10,000-plus computers that store and serve up its Web index. While the much-written-about perks—free gourmet lunches and dinners, a cafeteria full of cereal bins and Gummi bears, a sauna, subsidized massages—may seem reminiscent of dot-com bygones, Google doesn’t care. Its website proudly proclaims, “Google’s founders . . . built a company around the idea that work should be challenging and the challenge should be fun.” It’s a nerd utopia, and it works. The parking lot is usually half full until well after dark—that is, when not emptied for the twice-weekly roller-hockey games.
The company isn’t averse to making money. Its revenues for 2001, as estimated by The Wall Street Journal , were more than $60 million, about half derived from advertising. Google’s philosophy on advertising hews closely to the founders’ technological roots, namely: Ads are okay only if they do not detract from the user’s experience. Thus, Google sells two kinds of keyword-specific ads in a clearly labeled, text-only format to small businesses and big sponsors such as Sony, Amazon.com, Disney, and General Motors.
Page and Brin—now the presidents of products and technology, respectively—are determined to keep those ads from alienating people who use the site. As they wrote in an appendix to their 1998 paper “The Anatomy of a Large-Scale Hypertextual Web Search Engine”: “The goals of the advertising business model do not always correspond to providing quality search to users. . . . [W]e expect that advertising funded search engines will be inherently biased towards the advertisers and away from the needs of the consumers.”
Google has been supplementing its ad revenues by hosting both Web and single-site search services for the likes of Cisco Systems, Washingtonpost.com, MarthaStewart.com, and Procter & Gamble. Competing with FAST, Inktomi, AltaVista, and other outfits with similar offerings, Google basically helps visitors to these corporate websites find information as quickly and easily as they do at www.google.com.
The real money lies in taking the company’s search capabilities into corporate in-house networks, or intranets. But this will be tough. Companies don’t want to let Google—or any other outsider—through their firewalls. Nor do they like the idea of outsiders operating their intranets. Instead, says Avi Rappoport, who runs the search-engine consulting service SearchTools.com, “it would make sense for Google to look into productizing their engine.” In fact, Google has done just that. It has developed a software-and-hardware package, the Google Search Appliance, that was released in February.
This new product is essentially identical to one of the computers the company uses to index the Web, and runs the same software. There’s a $20,000 box that can handle an intranet of up to 150,000 documents, and a $250,000 eight-box rack that can deal with millions of documents. But searching an intranet is not quite as simple as searching the Web. For example, access levels must be set so that not everyone can gain entrance to sensitive human-resources information. And intranet documents aren’t linked in the same way as Web pages. It remains to be seen whether Google’s technology can handle these requirements.
Hardware is an entirely new area for Google and has its own headaches and concerns, like whether customers will be vigilant about updating the software, says Laura Ramos, a research director at Gige Information Group who’s seen the new package demonstrated. She wonders whether Google can build enough boxes and manage the software challenge effectively enough to make money.
Potential investors are sure to see the move into corporate search services as a sign of maturity, like Google’s hiring of Schmidt away from Novell last year, first as chairman and then as CEO too. One of the most immediate tasks on the technology veteran’s to-do list: find a chief financial officer; Google has never had one. “From an outside perspective, they have a gray-haired enough CEO in place now. But you need a CFO to IPO,” says Marc Baum, CEO of the IPO.com news site. “We’re back to the old world, where management is an important part of the equation.” As of March, the company hadn’t appointed a CFO.
Google remains an outfit backed by venture capital. Between them, Kleiner Perkins and Sequoia have brought to market all the giants of the Internet—Amazon.com, America Online, Yahoo!, eBay—as well as many companies whose bubbles have burst. The two VC firms are believed to have invested an additional $10 million in Google in late 2000. How much longer will they wait to get their investment back? “As much as they want to recoup their money, they don’t want to be associated with busted IPOs,” says Baum. “Google’s timing is going to be more about the equity markets.”
The unofficial consensus is that Google will probably go public by the end of this year. “They’re not worried about an IPO. They’re profitable; they still have money in the bank,” says Ramos. “It’s more important to the founders to build the best dang search engine in the world than it is to grow huge and make a lot of money.”
Brin and Page aren’t the only ones who feel that way. In a New York Times roundtable discussion held before he took the CEO post, Schmidt said, “What I always worry about is when the business seems to be about marketing to its shareholders rather than to its customers. . . . I hope we go back to a model where the product that we’re selling is the product that the company makes, whatever it is, as opposed to the CEO, the brand, the executives, the shareholders, whatever.”
He’s landed in the right place for that. And if the new Google Search Appliance becomes anywhere near as popular as the Web search engine, Google can afford to wait patiently until the perfect IPO window opens. “Mind you, the market is much smaller, and investors have just come out of the Internet bubble,” analyst Safa Rashtchy says. “But it is going to be one of the strongest IPOs that we will see this year.”
In the meantime, Google is putting the Web at our fingertips—free. Pass the Gummi bears.
