What’s the deal with Internet companies these days?
Deals! Lots of deals: High-priced acquisitions of fast-growing startups. Marketing alliances between old-media giants and new-media innovators. Distribution partnerships between companies that sell stuff and Web sites that attract eyeballs. It’s a fact of life in the new world of business: Companies live and die by how well they do deals. And their leaders have to cut make-or-break deals at speeds that would paralyze their counterparts in slower-moving industries.
“There are two options for Internet companies,” argues Kevin O’Connor, 38, cofounder and CEO of DoubleClick, the Web’s advertising superpower. “Get bigger fast, or get smaller fast. If you’re not one of the top three players in any category, eventually you’ll be eaten — or die. That’s a pretty good incentive to find partners that can help you grow.”
“Everyone on the Net is scrambling to get big fast,” agrees Bob Davis, 42, president and CEO of Lycos, a Web portal that recently signed a high-profile — and decidedly controversial — deal with USA Networks to create an e-commerce giant. “We did five acquisitions and three strategic investments in less than a year. There’s a clear sense that speed is a necessity. This isn’t just about building companies. It’s about laying the foundation for an industry.”
In some sense, doing business has always been about doing deals. And many of the old rules of deal making — about knowing if you can trust the folks across the table, about preparing for a negotiating session — still apply. But there are new rules as well. Internet deals get done in weeks (and sometimes even in days), and negotiations are just as likely to take place outdoors — perhaps after a game of ultimate Frisbee — as they are in a conference room. More and more negotiations take place through email rather than in person. And because so many Internet companies have neither tangible assets nor an established track record of success, the human factor looms larger than it ever has before.
“If you’re buying a television station, you’re buying an awful lot of fixed assets and a lot of history and predictability,” explains Bert Ellis, 45, chairman and CEO of iXL Enterprises, an Internet-service provider in Atlanta. Over the past few years, Ellis has been a single-minded acquirer of small Web-development agencies. Back in the early 1990s, he was a serial purchaser of television and radio stations. (He later sold his group of 15 media properties for $745 million.) And he sees a big difference between those two industries: “In the Internet business, it’s much more difficult to assess value. All we’re buying is people.”
Of course, as more deals get done, and as those deals get done faster than ever, more suspicions get raised as well: There is a big difference between making a deal and issuing a press release. “Pressware” is to the Internet business what “vaporware” is to the software business. “We don’t do ‘Barney deals’ here,” insists Bill Nussey, 33, president of iXL. “That’s what we call it when you announce a partnership that’s only about ‘I love you, you love me.’ ”
“We’re looking for properties and technologies with high rates of growth,” says Bob Davis, whose bookshelves are lined with baseball caps that commemorate past deals. “That’s why we’ve acquired companies like Tripod and invested in companies like Bidder’s Edge. The opportunities come at us in a fire hose, and the trick is to know what to pursue and what to decline. What’s clear is that you need to make deal making a core competency.”
O’Connor, Davis, Ellis, and Nussey are writing a new rule book for how deals get done, for how deal makers do their job — and for what happens after the deal. Fast Company turned to them and other elite Internet deal makers to create the ultimate guide to Internet deals. Are you ready to deal?
I. How to Buy on Instinct
From the day he came on board at Lycos, Tom Guilfoile, 34, knew that he was in for a wild ride. He had been a senior manager at Ernst & Young’s entrepreneurial-services group, one of the giant accounting firm’s nimblest divisions. But that was nothing compared with life at one of the Net’s fastest-moving portals. Guilfoile started work as Lycos’s controller in February 1996. By April, he had helped the company to navigate SEC regulations and to float an IPO. That IPO took place just 10 months after Lycos was founded, making Lycos the youngest company in NASDAQ history to go public.
And the momentum just kept increasing. “I remember calling my wife one night and telling her I had to work late,” says Guilfoile, who is now vice president of finance and administration at Lycos. “Around 11 p.m., Bob [Davis] called me and said, ‘I want to buy a piece of PlanetAll [an online calendar and contact-management system], and we have to get it done by tomorrow morning. There were other investors who wanted in. Bob liked the concept, could see the value, and said, ‘Let’s do it,’ without much tire kicking. I got an email at midnight with PlanetAll’s proposal, and by 3:30 a.m., we had a letter of intent to buy 10% of the company. It was on Bob’s desk by 9 a.m., and it had been signed by both parties by 11 a.m. I called my wife when we were done and said, ‘Since the last time I talked to you, we bought 10% of a company.’ ” (Late last year, Amazon.com snapped up all of PlanetAll for $100 million — which meant that Lycos had parlayed its investment into a $10 million gain.)
Guilfoile, a die-hard baseball fan, has a gray metal locker in his office, along with a photo of himself as a 13-year-old batboy for the Pittsburgh Pirates. To plunge into the world of Internet deals, he had to shed some of his CPA conservatism. “My inclination had always been to beat things to death, to analyze everything to the nth degree,” he says. “But in this industry, you’ll be swallowed up if you do that.”
Guilfoile and Dennis Ciccone, 48, Lycos’s vice president of mergers and acquisitions, have developed their own style of doing deals. They work quickly, and they trust their intuition. Last year, Lycos bought five companies — including Tripod, a popular Web destination for college students, and Wired Digital, the online offshoot of Wired magazine. As a result, the company now reaches nearly as many Internet users — 49% — as Yahoo! does, and it has been growing at a faster rate than Yahoo!
“With most of these deals, there’s not a lot to hang your hat on in the way of financial analysis,” Guilfoile says. “There’s no stable earnings growth to project into the future. You’re not buying oil wells or factories. You’re buying an opportunity, eyeballs, and people. That forces you to become a good reader of a company’s people and its culture. You’re looking for people who are committed and hard-working, who have the same vision and goals as you do. It’s not going to work if you’re dealing with people who are looking to check out and who see you as their ticket to retirement. The people you deal with have to be builders. Basically, we rely a lot more on instinct than on facts, because facts just aren’t available in most cases.”
Lycos is always on the lookout for its next acquisition, its next investment opportunity. And manning the lookout post is Dennis Ciccone’s full-time job. Ciccone came to Lycos after that company acquired his company, WiseWire, in April 1998. (That acquisition, which involved a lot of office sleepovers on the part of people from both companies, was completed in just one month.) There’s no shortage of work, says Ciccone: “I always have 40 to 60 business plans on my desk, and I’m on the road a few times each month, doing tours of promising young companies.”
Ciccone has recently focused on e-commerce companies that are developing shopping bots, tools for building Web storefronts, or software to handle back-end transactions. He relies heavily on input from product managers and engineers at Lycos, who provide him with blueprints of their strategy and who advise him on which technologies Lycos should buy rather than build in-house. “Evaluating technology is important, and it requires a lot of input,” he says. “We have to ask, ‘Does it work, and will it still work when we plug it into the Lycos network and pour on the hits?’ ” The e-commerce tour has taken him to Toronto and Amsterdam, as well as various cities in California.
Ciccone looks tired as he leans against a credenza in Guilfoile’s office. Because the expectations of Lycos’s users change so quickly and because the competitive landscape shifts at an equally fantastic rate, Ciccone’s universe of potential acquisitions is nearly infinite. “We’re not at a stage where we’re comfortable with blowing any one off,” he says. “You don’t want to miss an opportunity. If people email me, I email them back. If they call, I call them back. I can’t afford not to, and usually just one phone call can clarify whether or not there’s something there.”
Bo Peabody, another executive from an acquired company, operates a sort of deal-making skunk works at Lycos. Peabody, 27, came to Lycos in February 1998, after it acquired Tripod, the Web community that he founded in 1992. Lycos, which paid $58 million for Tripod, has given him free rein to seek out tiny, low-profile Internet companies and to shepherd them through the acquisition process. While Ciccone concentrates on big-ticket deals, Peabody finds companies that Lycos can buy for less than $10 million. Such deals are the kind that competitors tend to overlook.
“There are a lot of $5 million and $10 million companies out there that are looking to hitch their wagon to something larger, but they don’t get the attention of a Yahoo! or an @Home or an Amazon,” says Peabody. “Often they have a technology that leads you to say, ‘We could build this in six months.’ But in our world, six months is an eternity.”
Working with Ethan Zuckerman, 26, a technology specialist who also came to Lycos as part of the Tripod acquisition, Peabody appeals to the founders of Internet startups in a way that Ciccone can’t appeal to them. “We don’t use an investment banker, and we don’t come in with spreadsheets,” Peabody explains. “We talk about code. We talk about products. It’s deprofessionalized. It’s techie-to-techie, entrepreneur-to-entrepreneur. I know what it’s like to have a company with five people.”
One representative acquisition involved GuestWorld, the company that created the most widely used software for putting guestbooks on personal home pages. With 800,000 registered members, GuestWorld was clearly the leader in its category, and Lycos bought the company last June for $3.9 million. (Like most of Lycos’s deals, this one was an all-stock transaction.) “We did the GuestWorld acquisition inside of a week, from first contact to term sheet,” crows Peabody. “Doing that kind of a deal helps us move faster. I think some of the best value that we’ll get will come from those little deals — from being crafty and street-smart.”
Once Lycos starts negotiating, it blocks out as much outside interference as possible. “To get the other company’s undivided attention, we like to sign a letter of intent as soon as we can,” says Guilfoile. “And we ask for a 30- or 60-day period in which to sit down and do our due diligence.” In many cases, though, once Lycos shows an interest in a company, other portal sites declare their interest — which in turn encourages the object of Lycos’s affection to start shopping around. “We’re big fans of handing out offers with a fuse attached: If you don’t commit within 12 hours, this deal blows up,” says Peabody. “We want each deal to happen fast. We don’t want to give the other company time to talk with our competitors.”
II. How to Make a “Standard” Deal
Back in 1995 — ancient history by Web standards — streaming audio was simply the latest cool idea to hit the Net. Seattle-based RealNetworks (known back then as Progressive Networks) introduced the first version of its RealPlayer, a tool that lets Internet users listen to speech and music on their computer without having to wait for a huge file to download.
From the start, though, RealNetworks understood that its undeniably nifty technology would wither if it didn’t sign deals with strong content and distribution partners. When the technology was launched, the companies that supplied programming for it included National Public Radio and ABC News. Today 85% of Web sites that use streaming media do so via RealAudio or RealVideo. Among those content providers are more than 3,000 radio stations as well as hundreds of television broadcasters. By the end of last year, there were more than 48 million registered users of RealNetworks technology, and that technology had emerged as a de facto standard.
How do you make your technology a standard? By making lots of deals. “We’re smack-dab in the middle of building the Internet into the next mass medium,” says Len Jordan, 33, RealNetworks’s senior vice president of media systems. “Deal-making skills are an important commodity here. You have to partner with a lot more companies than I would ever have imagined. You have to be sharp in understanding what your company’s core assets are — and what its core desires are. You have to be efficient in matching your assets and desires with those of your partners, and you can’t be shy about moving aggressively to make things happen.” And on the Net, Jordan adds, you never know where your most valuable partners will turn up. “If you’re DaimlerChrysler, there’s not going to be another car manufacturer tomorrow that you have to know about,” says Jordan. “Because the capital requirements are so high, you’d know about any competitor years ahead of time. But on the Internet, a bunch of college kids in Illinois can come up with Mosaic, and that company can turn into Netscape in a very short time. In our industry, you can’t ever say that you know who all the players are, because new players are emerging all the time.”
That’s why Jordan has developed an internal radar to scope out potential partners. He is a devoted user of online news services like TechWeb, CNet, and Quicken.com, and he sifts through a half-dozen trade publications a week. When he first heard about Inktomi Corp., for example, Jordan was intrigued by the company’s caching technology, which makes files more accessible to users by storing copies in several locations across the Internet. “We instantly saw that their [Inktomi’s] technology could reduce the cost of streaming media, and we struck up a relationship with them. They were a young company at the time, pre-IPO, and knowing about them was just a matter of having our ears pricked up.”
Companies that hope to do deals in quantity also need to send clear signals — to their own people and to the outside world — that they are an open system. “You have to immerse your employees in the notion that you win by working with other companies, not by going it alone,” says Jordan. “And people outside the company need to know that the welcome mat is out. We take meetings with a lot of little companies — outfits with just two or three people — because it wasn’t that long ago that we were that size. You need to cast a wide net, because this is a big ocean. You can’t count anyone out.”
“Anyone” includes even Microsoft, which produces its own streaming-media player and which is RealNetworks’s foremost competitor. At one point, Microsoft had invested $30 million in RealNetworks, but the software giant divested itself last year, and the two companies are now battling over technology standards. Still, in the Internet business, a company must never say never.
“MSNBC is the only major news site that doesn’t use either RealAudio or RealVideo,” says Mark Hall, 35, general manager of media publishing at RealNetworks. “But we would definitely be prepared to take on the MSNBC site as a content partner. It has some great programming.”
Hall oversees many of RealNetworks’s content partnerships. “You want to have a system that’s scalable and efficient, because you’re dealing with tens of thousands of sites,” he says. “The simpler you keep the partnership, the faster lawyers can get through it.” Indeed, the standard agreement between RealNetworks and one of its content partners is only three pages long. Under this agreement, no money changes hands: RealNetworks provides technology, offers support services, and helps distribute programming to its audience of 55 million; the partner simply agrees to provide a channel of daily content in the RealNetworks format.
“We’ve done deals in two or three days, and usually we do them in no more than two or three weeks,” Hall boasts. Unlike the iron-clad agreements of the old economy, though, RealNetworks’s content partnerships seldom remain unchanged for longer than two years — and, in fact, one year is the norm. “You don’t want deals that prevent you from being flexible and nimble,” says Hall. “Change is what the Net is all about.”
Avoiding long-term contracts doesn’t mean becoming a shortsighted company. Maria Cantwell, 40, senior vice president of the consumer and e-commerce division at RealNetworks, agrees that lengthy and exclusive contracts often work against a deal. But, she argues, nimbleness and flexibility shouldn’t keep you from sharing your vision of the industry’s future with your partners.
“You need to identify opportunities in which you can build alliances that involve more than just one deal,” says Cantwell, who served as a member of Congress from Washington State before she joined RealNetworks. “For example, we try to think, What are the long-term benefits if America Online and RealNetworks work together? Will a deal with AOL help us build the market for streaming media? How do we work together in the future, now that all 15 million AOL subscribers can easily gain access to RealPlayer?”
Cantwell is fond of comparing deal making on the Internet to what takes place in the halls of Congress: “In the technology world, you benefit from cutting to the chase — being up-front about your common interests. Email and the Web let you share information and find common ground quickly. Technology accelerates the exploratory phase. Once two companies decide that they’re in sync, people here really want to do deals. In politics, though, you never know when you’re going to need people again, so you focus on building coalitions, building trust, and then working out the details. You’re also more fearful of a failed negotiation than you are in the Internet space. Politics is such a public arena: You’re always thinking, What will the media say?”
III. What’s the Deal with Talent?
Bert Ellis, chairman and CEO of iXL, oversaw a mind-blowing 34 acquisitions from 1996 to 1998. Bill Nussey, president of iXL, took part in about 20 of those deals. But the two men deny that they’re pursuing an old-fashioned roll-up strategy (one that involves trying to dominate a decentralized industry by assembling a bunch of far-flung assets). Instead, they’re simply trying to build the Web’s premier design, development, and engineering firm. “This isn’t a financial roll-up,” Nussey claims. “This is a talent roll-up.” The goal behind iXL’s acquisitions, he says, is to build a big team of talented people — and to build it fast. “Our acquisitions are a way of hiring a lot of people, while reducing the risk that they won’t be able to work together. We know that they can work together — they’ve already done it.”
Because Nussey is buying not hard assets, such as a factory, or even intellectual capital, such as a patent, but rather groups of people who know how to build things on the Web, his top concern in evaluating a deal is the human dimension. “We spend at least as much time looking at the cultural fit as we spend reviewing the books,” he says. “If we discover reasons why our culture and the other company’s culture wouldn’t mesh, we break off discussions.”
When deals do go forward, iXL makes sure that all of the acquired company’s stockholders have an incentive to stay on board: Selling your company to iXL is not an exit strategy. Shareholders are paid with iXL stock (which is not yet publicly traded — although iXL did recently register for an IPO). “The only people who will take our deal are those who want to be with us over the long term,” says Nussey. “We hope to attract people who love their work, but who want to be part of a bigger company and have a bigger impact.”
Since iXL aims with every deal to gain a fresh infusion of top-notch talent, it also works hard to make sure that rank-and-file people don’t jump ship. “If you have a conquest mentality, you’ll fail,” says Dave Clauson, 44, iXL’s executive vice president of worldwide marketing. Executives from iXL headquarters spend as long as three months working on-site with acquired companies. “One of our objectives is to ascertain how we can add immediate value. What is this company struggling with? What is it challenged by? Maybe there are technology problems that we can help resolve, or new-client meetings that we can help with. We try to generate early momentum by fixing some things and winning new business. That eliminates a lot of anxiety.”
IXL also holds monthly summits that bring together staff from each of its 14 U.S. and 5 European offices. One summit might bring together all iXL employees who are responsible for recruitment. Another might focus on helping salespeople share selling strategies. A third might serve as a conclave for creatives. Each summit is held in a city where iXL has operations. “Most companies underinvest in knowledge transfer and in culture and emotional connection,” says Clauson. “But that kind of investment is part of our battle to attract and retain the best people.”
So far, iXL is winning that battle. Of the roughly 100 senior executives who came to iXL from acquired companies, only 2 have quit. “We’ve become a sort of magnet for people in this industry,” says Nussey. “We’re on the map now. Because of our reputation, we’re being approached day and night, and our ability to acquire good people is increasing exponentially.”
IV. Dealing Your Way Back from the Brink
Companies that live by the deal can also die by the deal. Novell Inc., the once-thriving network-software company based in Provo, Utah, almost killed itself after signing a barrage of bad deals. At its high point, Novell controlled more than 70% of the market for networking software, and its leaders believed that they were building a credible competitor to Microsoft. So Novell embarked on an acquisition binge, buying 17 companies (including WordPerfect) in the span of a little more than two years. But the acquired companies diluted Novell’s strategy, which had focused exclusively on computer networking. Those companies also proved hard to integrate into Novell’s culture.
“I can’t think of a single acquisition in which the relationships really worked,” admits Christopher Stone, 41, now Novell’s senior vice president of corporate strategic development. “The revenues of the company shrank from $2.5 billion to $1.1 billion, and we wound up selling off most of the companies we had acquired.”
For a while, Novell looked like buyout bait itself. Then it sold off WordPerfect, installed a new CEO — and changed its deal-making philosophy. Forget gobbling up disparate companies and trying to become a Microsoft-sized behemoth. Novell’s deals now focus on building a product line around Novell Directory Services (NDS), which Stone calls “411 for the Internet.” NDS gives system administrators additional information about their corporate networks; it lets them identify and correct problems from any terminal connected to the Net; and it allows them to permit or deny users access to sensitive documents — a feature that helps to enable secure commercial transactions. For such a directory to have real value, it has to be used by all Internet-service providers, and it has to be built into the switches that control Internet traffic. “The directory has to be ubiquitous,” says Stone. And making it ubiquitous means lots of deal making.
For example, Stone put together a deal with Lucent Technologies, one of the world’s leading makers of switching equipment. Last October, Lucent announced that, starting this year, it would include NDS with its own software. Wall Street and the media paid attention, and a flurry of deals with other top-tier players followed: Cisco Systems, Nortel Networks, Tivoli Systems. “One of the rules of Internet deal making,” says Stone, “is that if you get the big fish first, everyone else falls in line.”
Sometimes it takes clever bait to land the big fish. At first, Cisco wasn’t very interested in NDS, Stone says: “So we went to all of our customers who owned Cisco routers, and we asked them to help us get Cisco’s attention. They set up a forum and started talking about how great it would be if Cisco’s products supported NDS. We got 500 emails the first day, and some of them were sent to [Cisco CEO] John Chambers as well. Then he started getting phone calls too.” Last November, Chambers finally gave in.
To keep its deal-making campaign moving at a fast clip, Novell used the Lucent contract as a template for future deals. “Developing the Lucent contract was a nightmare,” says Stone, adding that the process lasted about four months. “Everybody and his brother — every executive, every product manager, every engineer — had to have a hand in it. But having that contract as a platform made later deals much easier.”
Making sure that deals are quick and simple is important to Stone, who wants Novell to regain its reputation as a company that stays ahead of the pack. That’s why he does as much negotiating as possible through email and the Web, and why he does his best to keep money out of the discussion. “It used to be that technology deals were all about the licensing fee that I got from you in exchange for my software,” says Stone, who started and sold his own company — Object Management Group — before joining Novell. “The problem with that model is that I didn’t care how many people actually used the software; I only cared about the fees that I was collecting. In the Internet era, it’s all about installed base: How many people use the product today? Besides, whenever money is placed on the table, something gets screwed up.”
So most of Novell’s deals are joint marketing-and-sales agreements in which little or no money changes hands. That model is becoming more and more typical of Internet deal making: “These deals are about ‘I’ll mention you in speeches, use my Web site to distribute your product, and help you publicize the product,’ ” says Stone. “There are lots of things to focus on that are more important than trading dollars — like ‘Will customers buy it?’ ”
Stone is also a fanatic about paper: He hates it. “I won’t allow people to do anything on paper — except the final contract,” he says, sounding as if he can’t wait for the day when legal signatures can be rendered in pixels. Entrepreneurs who apply for financing from the Novell Internet Equity Fund fill out a Web-based form. Using keywords drawn from the form, a computer then forwards the application to an appropriate executive in Stone’s group. Discussions about whether to do a particular deal take place largely via email. Documents are edited online — which reduces the chance that an outdated copy of a contract will remain in circulation.
For Novell, doing deals digitally means identifying service providers — such as lawyers and investment bankers — who are willing to forsake paper. “It’s very simple,” says Stone. “You play by my rules, or I don’t hire you. Working electronically is just more efficient.” Many of Novell’s Internet investments — which range in value from $1 million to $3 million — are wrapped up in as little as two weeks’ time. The deals are templated (“all that changes are the names and the numbers,” Stone says), and Novell’s attorneys, along with a valuation expert from an investment company, review each contract via the Web.
Novell is also vigilant about never announcing a deal before it’s finalized. Stone believes that by putting out “pressware” — instead of releasing real software — companies erode their reputation and hence the perceived value of partnering with them: “People stop listening to you, because they assume that you’re crying wolf.”
And, Stone emphasizes, when Novell does issue a press release about a new partnership, that just marks the beginning of the deal. “You’ll really piss off your partners if you don’t follow through on what you promised,” he says. To enable proper follow-through, Novell assigns dedicated personnel to work with each partner. The company also holds monthly status meetings. At those meetings, Novell wants to know not only how each partnership is holding up but also how it can be expanded.
Novell treasures its relationships with partners because it knows that they hold the key to the company’s future. Novell’s people also acknowledge that Internet deal making — like all deal making — is social. That’s why, last January, Stone and Steve Adelman, 42, vice president of corporate development, organized the first-ever Novell Global Partners Summit in Snowbird, Utah. The event attracted 400 people. “It was an opportunity to get together for a couple days to talk about how we can work together better,” Adelman explains. “We want to leverage what we have as a network of companies, rather than as pairs of partners.”
And lately the news about Novell has made a sharp turn for the better. The company’s share price is up, and its marketplace momentum is back. “We’ve realized that Novell needs friends to be successful,” says Stone. “Things change so fast that we can’t do it all alone. In the beginning, we were nervous because of our long legacy of bad deals. But we feel that we have the formula down now. Doing smart deals is our mantra.”
Sidebar: How Steve Jurvetson Does Deals
Steve Jurvetson, 32, likes to boast that his firm, Draper Fisher Jurvetson (DFJ), has financed more Web companies in the past six years than any other independent venture-capital outfit. Many of the deals have been blockbusters. Microsoft bought Hotmail, a DFJ company, for $400 million. Yahoo! bought Four11 for $95 million. CyberMedia Inc. went public — and was then bought by Network Associates for $130 million. “We don’t sit around doing lots of deep thinking,” Jurvetson says. “We get a sense of what will be a billion-dollar opportunity, and we look for companies. We get to closure before others see that opportunity.”
Here are Steve Jurvetson’s rules for Internet deal making.
1. Get fast or get lost. “At most venture-capital firms, partners gather for a Monday meeting to review the business plans that they’ve received and to make decisions on how to proceed. We don’t wait until Monday. We make decisions on an ad hoc basis, and we take iterative steps throughout the week — in person and by email. We’ve also engineered our due-diligence process to be minimal. You’ve got to close deals quickly, or else you’ll miss out.”
2. Bigger is better. “The ‘800-pound gorilla’ in a category tends to get the dominant share of business and financial partnerships. Many advertisers and media companies don’t want to spend time with small properties. And that makes it tough for new entrants.
“Hotmail, for example, was doubling in size each month, but it took six months to reach 1 million users. Until it reached that point, the company was off the industry’s radar screen. By the time people came to realize that free, Web-based email was indeed a hot idea, Hotmail was adding 1 million new subscribers a month.”
3. Mix work with play. “Lots of deals happen in situations that aren’t ‘pure work.’ We play a weekl