Vinod Khosla Talks Shop

(page 2 of 2)

That's why there's all this interest in bio-fuels, because it is cost effective and they can scale without subsidies. Today, ethanol could scale without subsidies. Big oil is trying to make sure E85 doesn't happen. [E85 is a blend of 85% ethanol and 15% gasoline that is cheaper than gasoline to produce.] They're trying to preserve their segment.

As a successful entrepreneur yourself, does it get easier after the first time?

Yes. You learn about how to do it and you learn how to scale. You learn what not to do and what you don't know. I don't worry about scaling too much now, because if we have a technology breakthrough, we know how to hire the management that knows how to scale. We are always looking for both the technologists and the management teams.

What were the biggest mistakes you made the first time that others should avoid?

The hard part is all the stuff you don't know you don't know. That will happen no matter what. Even the basic idea of what scaling is about—the way to implement scaling. When do you worry about it, when don't you? What's scalable, what's not? What's too hard? How do you fit into others' expectations and leverage the ecosystem? How do you minimize your liabilities and maximize your assets? How do you get funded for subsequent rounds of financings? How do you hire the right teams? Make corporate partnerships? It goes on and on.

What are the areas that are ripest for entrepreneurs to get into?

I find the very notion of "hot areas" counterproductive, because hot to me means an over-invested area. In 2001, I made a brand new, very expensive, optical telecom investment when it was the worst area. In some sense, I did that because it was the worst area. And guess what? The company is doing very well. In the first six months of shipping, it did $25 million, and this year, they'll probably do $100 million. That's explosive growth in an area that wasn't very hot.

On the other hand, the basic mobile infrastructure space was best for large companies that could bring a large amount of money to it. Startups do well when they develop a surprising new application, and today, mobile is a ripe space for startup innovation and much less amenable to large company processes. The intersection of Internet, Web 2.0, and mobile is an exciting area for venture capital.

You've got degrees in engineering from Carnegie Mellon and business from Stanford. To what extent is it important for entrepreneurs to have different disciplines in their background?

I don't think it's essential, but it clearly helps. Engineering's thought processes are very applicable to non-engineering areas. Having said that, I think too mechanical a mindset will also hurt. So, as usual, it's a happy compromise between multiple points of view.

I'm always a big fan of what I call "managed conflict." That means getting people with different points of view together where they don't always agree because they have different backgrounds and different assumptions about how things are usually done, the truisms in their business.

The best ideas come out of healthy debate around established truisms and their intersection. When starting a company, I tell people to engineer the gene pool of the people to have different perspectives, different backgrounds, and intuition, and to let those different intuitions be in conflict, but in managed conflict.

That's why the management and the chief executive officer are so important. Their job is to assemble a great group and then manage them through this process. The best ideas come out of these kinds of debates. A startup isn't going to do well if it's doing what other people believe. It has to have a different point of view. A big company can follow conventional wisdom and do very well, startups have to have a unique angle. They have to change the rules in some way, believe in something others don't, and leverage that into a business strategy and business advantage. That's the process of a startup.

To what degree do you think there's a bubble in the social networking market?

You see good investing, and you see some herd investing in Web 2.0, and I think you see it in clean energy and mobile too. Whenever herd investing starts happening, be careful, because people want to be in a hot area, and you start to see inflated valuations. If I were an entrepreneur, I'd want a bubble too. But entrepreneurs can also get caught up in it and get hurt.

You'd want a bubble, but wouldn't you also want to get out at the right time?

You'd want to accumulate all the resources you can, but use them in a cautious way as an entrepreneur. But, if you're an investor, you'll get burned unless you're doing it intelligently. Even in the worst bubble, Google (GOOG) was still a great investment in 1999, right? Because the fundamentals were right and the entrepreneurs were right.

The other thing to keep in mind in these kinds of startups is that many more companies lose money than win, but much more money is made than lost. If there's 100 companies, 95 may lose money. But as a segment, it still may have a great rate of return. You see this in the mobile business, telecom business, biotech business. This is pretty typical of innovation cycles. In technology, it tends to be more "winner take all."

As an entrepreneur, you want the best help, not the cheapest money. You want to improve your probability of success and be one of those four or five winners, instead of wanting to take the money.

Jeffrey Gangemi is member of the MBA Class of 2009 at the Johnson Graduate School of Management at Cornell University. He was previously a staff writer for BusinessWeek.com.