|
|
|
 |
|
||||||||
|
|||||||||
|
|||||||||
|
|||||||||
 |
|
 |
|
|||||||
|
|
Top VCs scan valley, beyond for next hit By Matt Marshall Mercury News January 13, 2002 Venture Capitalists look past dismal 2001, discuss coming year As the economy shows signs of climbing out of the recession, where are venture capitalists in Silicon Valley putting their money this year? We talked to Promod Haque of Norwest Venture Partners and James Wei of Worldview Technology Partners, who were among the most successful VCs in the late 1990s. Like other VCs, they made most of their money over the past five years from communications revolution - which has since taken a beating. Q: Where are the best places to invest now? A: (HAQUE) Aside from communications, we're investing quite a bit in enterprise software. There is a move in the software industry to what's called Web services - delivering (mission-critical) applications over the Web. . . . Second, there's a gaping hole in the whole area of supply chain software. You look at Cisco, which wrote off $2.5 billion worth of inventory. Before the downturn, everyone, especially Cisco, was touting that they could close their books by the hour, and they've got a handle on their supply chain. They were $2.5 billion off. Finally, we're intrigued with the whole wireless (infrastructure) arena. Q: Is telecom still the VC's mainstay, though? A: (WEI) These are some great pieces of the communications industry to invest in. . . . Despite the Internet bubble (bursting), two related developments will continue. The first is the switchover from circuit switching to packet switching. The second is the compressions of protocol layers in the network. Q: Explain what you mean. A: (WEI) About 80 percent of what happens on the communications network today is done through circuit switching. When I call you, a circuit gets established between our phones. . . . The whole line (from end to end) is used, and anything above or below the frequency we're using is completely wasted. So carriers are moving to packet switching, breaking up the conversation into pieces, and bouncing them throughout the global network in whatever way is the most efficient. Second, the hodge-podge of protocols built over the years means that your conversation has to be switched through points in the network, requiring (redundant) equipment, space, management systems, cooling, and back-up systems in case it all fails. That's costly. So companies like Juniper, Cisco, ONI, Sycamore and Ciena are busy compressing protocols with next-generation networks. But they've installed only a tiny drop in the bucket in terms of (the network that needs to be laid). A: (HAQUE) The conversion from circuit to packet really hasn't taken place yet - it's yet to come. There are countries that are just starting that revolution. Look at China and India. If you look at where Cisco's sales gains are coming in from last quarter - it's based on was happening in China. There were billions, maybe a trillion dollars of investments in circuit switching (worldwide). Now it will take 20 years to upgrade it. Q: Which part of the network - specifically - can big money be made from? A: (HAQUE) Look at the whole data-center marketplace, which has evolved with the emergence of the Internet. . . . Now, the carriers are beginning to operate their own data centers. They're the best equipped to do this, because they have their own Internet backbones. However, the existing hardware and software is not optimized for data-center usage. This presents opportunities for a next generation of hardware and software companies (that build data centers). A: (WEI) The biggest shortterm opportunities might exist abroad. . . . Last week, I was in China, and it's amazing to listen to service providers talk about their architecture. They're putting data centers in the bottom of multitenant buildings. There's a ton of opportunity there. Q: So James, are you going to abandon the United States for China? A: (WEI) We don't invest in China because we still think the U.S. is the best place to invest, for cultural and regulatory reasons. There's the cultural side of building and managing companies. They're rapidly catching up on the entrepreneurial (front). . . . The regulatory environment is not very good for venture capitalists. It's hard to get your money out. You don't get the same level of transparency,or the legal protection, that you have in the U.S. But as a market opportunity, China is enormous. . . . That's why we at Worldview have business-development managers who help take U.S. technology companies to China. . . . They're building a new carrier once every four months. Q: So where's the best region for investing in the United States? A: (HAQUE) Silicon Valley, if anything, is going to become more attractive going forward. . . . Look at the technology infrastructure. For enterprise software, you have Oracle, the largest database company, based right here. You have the second-largest database company, Sybase. The two largest application-software companies in the world, PeopleSoft and Oracle, are based here. Ariba, Interwoven, Commerce One and Epiphany are all here. The list goes on. Then look at the data-networking companies. You've got Cisco, Juniper, Extreme and Foundry . . . and then a whole bunch of new ones that are coming up. Then you look at the telecom infrastructure, which traditionally has been a little weak in the Bay Area. But that's changing. Look at Juniper, which is working very much in the core of the network, and at Cisco's initiatives, and at a whole bunch of other start-ups. Q: What about the other regions around the United States that have boasted a high-tech revival, for example New York or Florida? A: (HAQUE) With the exception perhaps of Boston . . . you really can't think of a lot of other telecom companies outside Silicon Valley. A slew of dot-com start-ups, for example in New York and Florida, is what brought prominence to some of these other centers. . . . They're all out of business. One of the valley's biggest predicaments two years ago was, "Gosh, there's no space to grow here!" Well, there's 40 million square feet of rental space now available to anyone who wants to rent, so that problem's fixed. The money is here, the talent is here, and the space is here. The entrepreneurial climate is here. A: (WEI) Silicon Valley is getting stronger in another respect: The discipline is being put back. When the market softened, we wanted to reduce the burn rates of some companies we invested in. We had companies, for example, where the management team came back and said, "You know, we're going to need this many people to get the product out of the door. Rather than cut all these people, we'll take a salary cut." That's a sign of discipline. Q: When will things look up in the high-tech telecom world? A: (HAQUE) We're still in a stage where there are too many companies. We call it the "flea effect." This is a dog space, because there are too many fleas on this dog, too many me-too companies. On last count,there was 200 optical-networking companies. But there are not enough customers - the big telecom carriers - to test and buy their products. A: (WEI) Another problem is that prices that many carriers charge their users are declining faster than the costs. . . . It could be 18 to 24 months before we know whether some of the newer telecom service providers are going to make it or not. Q: So how do you pick winners, when some many are going to be losers in this environment? A: (HAQUE) The trick is to be the first to invest in a space. Don't go into a space where there's 60, or even six, players. A: (WEI) The key is to spend a lot of time with the customers who will buying from our portfolio companies. . . . and understand which technologies are important. We do this every month, and it helps understand which of the 200 optical networking companies will survive. And it's interesting: There are a few companies that constantly emerge - the ones that will save 20 percent in costs. They have to be fundamental painkillers for the carriers, save them lots of money and taking them to the next level of architecture at the same time. 
|
|
|
|
|
|
|
