Mike Volpi is senior vice president and general manager of the routing technology group at Cisco Systems Inc. Following Cisco"s annual analyst conference in San Jose this month, Volpi spoke with Computerworld about the changes the company has undergone since the bursting of the dot-com bubble. Excerpts follow.

How many companies has Cisco acquired in all? About 100, with 75 (of them) between 1996 and 2000. We were growing like gangbusters. When I started (in 1994), Cisco had about 1,600 people, and in 2000, we had 55,000 employees. Our revenues had gone all the way to US$23 billion. Some growth was acquisitions, and some was because of technology developed organically inside of Cisco. IT was just exploding.

And then the bubble burst. In 2001, things went south for the entire industry. Money disappeared, the bubble burst, people stopped spending, service providers went away. Our record quarter was $7.2 billion, and that dropped to $4 billion-plus.

We had layoffs, really tough -- almost 9,000 at once.

So what happened next? For a while, we were searching for bottom because there was this feeling of, "How far will we actually fall?" I felt we had gotten to the bottom by the summer of 2001. Then it became (an issue of) transforming the company, because every aspect of the company had been geared towards growth, and we didn"t have any great cost controls. Everything was very decentralized. Along with the layoffs, we did a restructuring where we recentralized a lot of functions. We decided to cut costs in marketing and engineering, exited several business areas and shut down product lines.

What are some of the technologies Cisco stopped offering? We exited a bunch of wireless areas, also DSL (and) long-haul optical. That was tough because all of those theoretically had promise. Looking back now, it might seem obvious why we exited areas. Why would anybody buy any more optical gear? you might say. But at that time, in 2001, we had no idea. You had to make these decisions amidst a lot of uncertainty.

There was a question in 2001 of whether we should even stay in the service provider business at all. Today, people will say, "What, are you insane?" But back then, it was a valid question. At the peak, before the downturn, close to 40 percent of our revenue was from service providers. And it had dropped down to 15 percent.

And it"s about 25 percent of your revenue now? Yes, we"ve been clawing our way back to a good position, and we can do more.

Do you think service providers will ever become so effective that they"ll be more trusted than internal IT is? If you talk about smaller businesses, they don"t have an IT guy -- and if they do, it"s probably a kid out of college. In those circumstances, service providers do have an opportunity to create value. But big businesses will continue to do IT on their own and simply buy connectivity.

What was the process that brought Cisco back, from an overall standpoint? There were a lot of changes in 2001. From there, it was just tighten your belt really hard and wait for the market -- weather the storm and see what"s going to happen. That was through 2003, and we made virtually no acquisitions, didn"t invest in new stuff hardly at all.

In 2003, we saw gradual improvements in the U.S. economy, and consumer confidence started going up, and so we started branching out a little more with a few acquisitions here and there, starting new product development, hiring a handful of people.

We have a better realization of what we can and cannot do as a company. Back in 2000, we all felt invincible. Now, we have the attitude that we know how to do certain things well and we don"t do these other things well, so let"s not do those things. We feel we still have to stay tight with belts because the climate"s not great, but we feel like "not great" can continue for a while and we"ll be OK.