Now, this sort of approach only works if you can empower the people who work for you. If your style is top-down and centralized, you'll get very consistent performance, but you'll also limit growth. If you're talking about keeping up with growth across as many areas as we are, you really have no choice but to empower. That's not only part of our culture. We also have the applications, the network, and the data design to let that occur.

And yet we have a healthy paranoia that makes [Intel chairman] Andy Grove look relaxed. Any company that thinks it's utterly unbeatable is already beaten. So when I begin to think we're getting a little bit too confident, you'll see me emphasizing the paranoia side. And then when I feel that there's a little bit too much fear and apprehension, I'll just jump back to the other side. My job is to keep those scales perfectly balanced.

How does Cisco's share price fit into this? For a long time, your stock was on such an amazing run that if you sneezed, 200 CEOs would come to San Jose to learn to sneeze like John Chambers. Now everyone's stock has taken a pounding -- including Cisco's. How much does that change the cachet of the whole company?

Well, there's a huge difference between the short-term stock performance and the overall strategy. And it's important to understand that. Our story of how we've evolved Cisco and how we continue to evolve it has never been stronger. In fact, I would say that if we do this right, you'll be writing an article 6 months from now, or 12 months from now -- whenever the economy turns around -- about how Cisco handled this in classic Cisco style and broke away again. It wouldn't be the first time we've done this. We did it in 1991, we did it in 1994, we did it in 1997, and we hope to do it in 2000-2001.

What builds good teams is found in how you handle stress. It's painful, but it's very good in terms of developing the character of an individual or a company. We want to show that we increased our investment in e-applications during this time period and used it as a chance to break away even faster.

Within Cisco, that makes a lot of sense. But surely some customers' appetite for working with Cisco goes up and down with your stock price, no?

Some CEOs will very definitely follow our main messages about the Internet. GE is an example. Others will say, "John, you don't understand. I've been CEO for two to three years already. The average life expectancy of a CEO is less than five years. If I don't make my profit target the next year, I've got a bigger issue." So I have to learn how to balance both of those perspectives.

We can't change how people behave. And we ought to listen to constructive criticism, because there are areas that we need to improve in. That said, we focus on customers who really believe that this is the future and who are willing to invest the time with us. We have 75 engagements with GE right now, for example. Others will say, "You know, I'm going to put this on the back burner." All you can do at that time is educate them and get them to realize that they are making a very conscious decision not to pay attention to the Internet.

When customers decide, for whatever reason, to change their spending behavior, we don't pack up our bags and go home. We stay closely related. They may come back to us later, either when the economy improves or when their perspective on the future changes. Over time, that's absolutely the right way to run your business.

How do you protect the Cisco brand in a downturn this severe? You've got an inventory overhang, fierce competition at the high end from Juniper, and a lot of speculation that there will be a price war whether you want one or not. What can you do to prevent brand erosion -- and market erosion?

What we're seeing in our industry is a rapid consolidation, where I believe the strong will get stronger and the weak may get eliminated. This is especially evident among companies such as alternate service providers, where I think brand will be more important than ever. I believe that during times of consolidation, Cisco is one of the best brands. Our market cap is greater than that of our eight largest competitors combined. We have $17 billion in cash, and we have balance among our product families, geographies, and lines of business, which will help us emerge stronger in the future. Brand is critical, and in the long run, it will be key to survival.

George Anders (ganders@fastcompany.com) is a Fast Company senior editor based in Silicon Valley.

Sidebar: What to Do When the Customer Isn't Buying

John Chambers has always listened to customers, even when the message hasn't been what he hoped to hear. Here are three reasons why CEOs are having second thoughts about the Internet -- and the responses that Chambers hopes will carry the day.