The one thing that is like death in an organization is passive aggressive behavior. Even the software transition we were going to drive, there were many hardware purists who said it's not going to work. And I'm like, that's okay. You can believe what you want to believe and I'm going to believe what I want to believe and you can be successful somewhere else.
Today on the CEO Signal, we're talking to Chuck Robbins, the chairman and CEO of Cisco. He took over a company that had helped define the early internet infrastructure but had missed the shift to the cloud. Since then, he's been repositioning Cisco for another epic infrastructure shift, this time around AI. From Seapore, this is the CEO Signal.
Morning, Penny. Great to be back together. How are you?
I'm doing well. How about you?
Good. Looking forward to this. We've got Chuck Robbins with us today from Cisco. He has been with this company since the 1990s. He's seen every shift in technology. And he was actually the very first CEO I chose to interview when we launched the CEO Signal newsletters about 18 months ago.
Really? Why did you do that?
Well, for two reasons. Because he has positioned Cisco to be at the very center of the AI revolution. It's a really interesting strategic shift he's pulled off here over the last 10 years or so. But secondly, he's also chairman of the Business Roundtable, which is the great gathering of senior CEOs in Washington, so he's got that sort of big view on corporate America too.
Yeah, I'm really interested in his perspective on AI as well as the business community at large. But frankly, one of the other things really on my mind is how did he take a hardware company and think about a major strategy shift before AI was obvious.
Yeah. And I think what I found interesting in my previous conversations with Chuck is he brings this measure of calm to that challenge which is not necessarily common in every CEO facing this much churn at this time. So I think it's going to be a really interesting conversation.
AI, AI, AI, AI, AI, II, AI. Isn't that great? We help cut through the AI noise so you can solve the business problems that matter most. Here's the plan. AI solutions from PWC.
Chuck, welcome. It's really great to have you here.
It's great to be here. Thanks for having me.
Thank you. Thanks for joining us, Chuck. I was telling Penny earlier, when we first met, what struck me most is you're in the middle of this AI swirl, you know, big sort of shift in the business you're trying to push through in your time as CEO. You're also chair of the Business Roundtable, not an easy role right now trying to navigate the relationship between corporate America and Washington, but you seem remarkably calm.
Well, when you're wrapping up year 11, you're a little calmer than I would be if it was year two as CEO. But I mean, what are you going to do? You just have to deal with the world and deal with life as it is.
Where'd you get that perspective from?
I'm sure it was growing up in rural Georgia with my family. But I've always had this, it's innate. I've always had this ability to, or I just don't worry about things that I can't control. We scenario plan for them. You have to plan for them. But you don't, you can't worry about them because there's nothing you can do about them. And I'm blessed that I have that because not everybody does. But it's not something I chose. It just sort of happened.
You run one of the defining Silicon Valley companies. You know, you got the Golden Gate Bridge in your logo. You don't come across as what we think of as the classic Silicon Valley type.
My accent give that away a little bit.
And maybe your suit and tie.
That's true. That's true.
Is that sort of having a bit of distance, is that a help or a hindrance in this job?
Well, I think that coming in with a different perspective matters. I mean, we've talked about diversity forever and, you know, today people are a little more afraid to talk about it, but I still think it's super important. I've always said if I had nine white guys from the South in a room with me, odds are we're going to align pretty quickly on what the answer is, but it may not be the right one. And so having people, I have a different perspective because of how my journey and where I came from, and the rest of my team does as well, and it just creates better decision-making.
Chuck, if you look at Cisco, let's say even just versus 5 years ago, there's been a lot of change in the company and you've really taken the company from what I remember as mostly a networking hardware company to one that really has a continuum of offering a real platform. Tell us about that transition.
Before I became CEO, I was running sales and I think in any given quarter we were taking orders for that quarter's revenue, like 85% of the revenue we were taking the order in the quarter. And that's a very stressful business model. You could be in week 11 of 13 weeks and think you're all set and then 12 and 13 don't work out and you missed the quarter. So that was one side of it. The second side of it, some of the changes we made were that at that time there was a meaningful shift to solving problems through software if you could. So we started, it was really you can deliver faster innovation if you're doing it with software as if you don't have to wait on custom hardware. But we all obviously understand the need for custom hardware. So what I was looking to do is build a balance. I would say we're going to transition to more software and more recurring revenue from a business model perspective, but we're also going to be one of the best hardware companies on the planet. And so fast forward today, 50% of our business is recurring from a much lower number back then. But the irony is if you look at what's driving our business today, it's the hardware side of it with all the AI buildouts that are occurring. So the great news is we have both and a lot of that was, there was obviously decisions we made along the way, but a lot of it was luck landing here.
I'd love you to take us inside that AI buildout story because I came to your AI summit in early 2025 this year. You had a lineup of the biggest names in AI. You know, Jensen Huang, Sam Altman, Fei-Fei Li, all these people are partners of Cisco now, customers. You got relationships with everybody in this kind of revolution if you like. How do you describe what Cisco's role is in that buildout?
So, everybody thinks about the GPUs and there's a lot of discussion about the GPUs with Nvidia and AMD and TPUs and Amazon has talked recently about how their success with their chips. At the end of the day, you have to connect all these things. And so, it's not the thing that everybody talks about, but it is truly the plumbing underneath is this connectivity layer, which is what we provide. And it's anchored in a decision we made almost a decade ago to buy a silicon company that, as it turns out, has become the platform for building networking silicon that's capable of actually supporting these training clusters. And the other role we play is in the enterprise. The enterprises are trying to figure out how to deploy this stuff and we're a trusted partner to enterprise customers around the world. So a lot of the players in the ecosystem are interested in partnering with us to make sure that they're optimizing their capabilities as part of a solution to the enterprise. And we're also deeply involved in securing the models. And you think about you're connecting but you're also really in a security business.
You're allowing companies to understand where they're vulnerable, to observe and to take action.
That's right. On the security front, the unique opportunity for us is that the network is where security has to be applied in the world we're in and moving to. So you think about this whole notion of agents. If I have autonomous agents operating in my infrastructure, I have to know what rights do they have, what data sources can they access, what are they allowed to do, who are they allowed to send emails to. I mean, all of these things, if you think about the identity security we do for humans, that's one thing. What we need to do for agents is going to be a multiple of that. And so the reality is you can only do that in real time. And the consistent layer is the network. And so we have a lot of software assets. We have Splunk. We have cloud security. We have all of these software assets that are important. It's a combination of both, but it's becoming more prominently a deep combination of security services running on the network today.
And is that a unique thing that Cisco is doing? Do you have a lot of competition in that?
Well, what I've said to our team is that none of our big security players have a networking business and none of our networking competitors have a security business.
So Cisco today is, you call it the infrastructure behind AI, and that wasn't necessarily obvious five or 10 years ago as the positioning. And in fact you've talked about missing the first wave of the hyperscaling for the cloud. So when did you know and did you know, I want to ask this question about uncertainty because things are changing so quickly. When did you know that the pieces you were putting together made sense and that it was going to fit the moment? Or is it something else?
Well, we had a lot of the pieces and we made the semiconductor, the silicon acquisition back in 2016, which I tried to convince people that I was quite the visionary, but I think that everybody knows that I'm not that smart. So we needed it and we bought it for different reasons. But as the AI movement began to emerge, we started to have more conversations and we began to realize that that was going to be a massive differentiator for us because the speed and the pace, if I'm buying silicon from the same source that all my competitors are buying it from, then I just look like everybody else. And so having our own silicon and the ability to actually have multiple silicon trains running at any given time with the next generation and the next generation and the next generation allows us to control our fate. And so that was a really big moment when we realized that we had an asset that was going to allow us to build this infrastructure. But then you have to execute because if you miss a cycle, you're out.
When you're in that phase, in that period between having identified the gap and the rewards from your decisions coming through, was there any pushback to things like the silicon acquisition?
The real reason we made the acquisition was to consolidate into a single architecture. And it was also like a six-year effort where you didn't get any return. So you're spending this money but by the time you get the silicon built, you get it spun, you get it taped out, you get it back, you get it called, then you get it inserted into products. You get the operating systems ported over to it. I mean, that first wave took a while. So you're making a bet without a lot of return.
What did it take to keep people inside the company and your shareholders on board when you're not getting a return for 6 years?
Well, the good news is that most people understood the vision where you just have to keep telling the story about where we're going. And to be fair, it wasn't a massive investment relative to the overall, but clearly those dollars, if I was looking to invest them for the return next quarter, the quarter after, I would have put them somewhere else. Today we're actually funding it as fast as we can, right? To continue to build out our capacity and, you know, now you're dealing with where do I find the talent? All the semiconductor engineers and product people are so in demand. It's like cyber security, you know, five, six years ago.
Did your background in sales, I mean obviously you're quite technical but also you're close to the customer coming from sales. Did that help you see the possibility or is really that the technical side was what was revealing where you needed to go?
It was a little of both because you referenced earlier that I've said on many occasions that we missed the first cloud transition. So when I became CEO and we started trying to rebuild those relationships, you had to listen a lot and listen a lot and you had to let them tell you how bad you were for a long time. And then you had to deliver. And so you had a combination of listening to what they wanted, understanding that many of these customers are so big, they're a market of one and you have to deal with them as a market of one, and then you had to deliver on the technology front. So it was a combination of both. The good news is I had super smart people on the technology side that were also great listeners and wanted to build partnerships with these customers and I think that really brought it together for us.
Was anybody recalcitrant? Did your board see where you were going? Did your sales team get it or the engineers or did everybody sort of see your vision similarly?
How'd you deal with that?
You get rid of them. I mean at the end of the day if you, the one thing that is like death in an organization is passive aggressive behavior. And so when you have someone who appears to agree with the strategy but then goes back to their team and said that's never going to work so we're not going to deal with it. You really do just have to make a decision that, and it's okay if you don't agree with me. And I would tell people that and it's like, it's fine. I get it. Even the software transition we were going to drive, there were many hardware purists who said it's not going to work. And I'm like, that's okay. You can believe what you want to believe and I'm going to believe what I want to believe and you can be successful somewhere else.
And was that a technical difference or is it a culture change too?
Yeah. Well, so Cisco grew up as a serious engineering culture. And so for years the way you won is you built the next 10 features. And you built the next 10 features and the next 10 features and at some point you're building features that 2% of the customers are using. And you're introducing in many cases more complexity because of all the things you're bringing to it. So it seemed pretty obvious to me that the market was moving towards the user experience even in networking. Like you had hardcore engineers that still wanted to work on command lines and these old prompt interfaces and things of that nature. That everything was starting to move faster and we needed to give our customers the ability to actually get to the outcome they were trying to get to faster. And that meant making it simpler and making it easier. And we needed to do that. It wasn't about the next 10 features anymore.
So once you've identified the need for that culture change, what do you think were the most important steps you took to make it actually happen throughout the organization?
Well, you got to get people who align, you know, conceptually with where what you want to do. And by the way, it wasn't like I walked in a room and said, 'Hey, this is what I believe. You guys get on board or you don't.' It was a team effort for us to, we all, we bounced it around and it shifted and changed and we adapted it. And then you have to just communicate, communicate, communicate, communicate. And I think that the old adage you have to tell somebody something three times, that's not true. It's many more times than that. So we have a big communications culture, very transparent, with our entire organization and you just have to, we get the whole company together once a month and we just go through this stuff over and over and over and just tell people where we're headed because what people try to figure out is they have a daily, weekly, quarterly pressure in their current job. And then what we're trying to get them to do is think about what's your role relative to where we're going two or three years. And that's hard for them to get to. That's the thing you have to try to figure out.
Yeah, I found communication particularly internally is one of the things as a CEO or chairman is just at the time that I'm exhausted from repeating myself is the time that people are beginning to understand what you're trying to tell them. And so you can't overcommunicate enough.
Going back to that period, what were the stakes? You know, if Cisco hadn't made these organizational, strategic, and cultural changes, what would have happened?
If we had not bought the silicon company back in 2016, we would not be sitting here having this conversation right now. We would be a company trying to figure out what's our role in AI. We're going to hopefully connect and we'll probably focus on security. But having the underpinning of the silicon is a massive advantage for us.
So, one of the biggest bets you made along the way was a $28 billion acquisition of Splunk. Huge move into security and visibility. What did you think that gave Cisco that you needed so urgently?
At the time the vision was that there's all of these insights that exist in your technology infrastructure today and there wasn't a way to consolidate all of it. So coming out of network devices, security devices, endpoints, email, everything and there were lots of individual assessments of a sector of that data but not a consolidated view. So the whole idea with Splunk was let's get all of that data back to a single data source where we can aggregate it and we can make decisions and customers can see what's going on. So that was the vision which in a normal world was reasonable, in an agentic world is required. And so again, it was a good decision and also a lucky one relative to where the market was going but our vision for what we saw across the infrastructure was the same thing with or without agents, it's just going to be magnified with agents.
