Interview of Everything: Cisco CEO Chambers on white boxes, SDN, leadership and the cloud
A while back you talked about your goal of becoming, I think the term you used was the most important IT company at that time.
No.1 is the way I defined it in terms of the importance to the customers. The No.1 IT company isn't by volume, it's in relation to business customers because those are my customers, not the consumer. Who do they view as their most important partner That's my definition of the No.1 IT company.
That was a big goal. Where do you stand
It was very similar if I were to draw a parallel to 1993, when we said we're going to change the way the world works, lives, plays and learns, and the Internet is going to be at the core of that. Everybody said that's really catchy marketing, John, but you know you're a router company. And yet, I think you would probably say we did that more than any other company. When we originally said we were going to become the No.1 IT company people said "Yeah, that's cool but unlikely."
And if you watch, every transition now is off the network. Almost every move in the market is either a move to align with where Cisco is going or to align to compete against us or to utilize that technology. So to say we were in the right spot at the right time would be an understatement and our strategy around architectures and how you tie this together to get business outcomes is working. I can't mention the companies, but you're seeing $100 million complete network upgrades. Not just to upgrade the network, but to position for the Internet of Everything, business transformation, security, etc.
The Internet of Everything took a little bit longer to take off than I thought because we started on this eight years ago and I had to buy people drinks then or even three years ago to get them to talk to me about it. Yet between CES of last year to this year, it went from being a surprise, to being everything this year.
At the World Economic Forum it was one topic a year ago on a panel and this time there were 21 panels on it. You don't need to explain to [French Prime Minister Manuel] Valls or [German Chancellor Angela] Merkel or [UK Prime Minister David]Cameron or [Indian Prime Minister Narendra] Modi what this means to the future of their country. Very often, if they think of one player that can help them achieve this goal, it's going to be Cisco. If you watch, just as an example, when companies are thinking about what can you do to align with Cisco and you saw us put our best foot forward in terms of why we think they should do that, it's really working and the size of our deals are the biggest they've ever been. But when countries do that, that's never occurred before and it gives you an idea how much of the tipping point we are. We've got a very good opportunity to achieve the goal. It's now about execution. Most people would give us pretty good marks in vision and strategy.
You've mentioned a 30% leadership change within Cisco over two years. What drove that and what were the outcomes you were trying to get to
This is why you always listen to your customers and your peers. What I've realized is most leaders cannot reinvent themselves at the CEO level or at the operational level. So as you make transformations at the speed that we're now doing, you have to probably change between 20% and 40% of your leaders and that's hard for me to say because, as you know, I'm very close to my leaders. For example, in engineering, as painful as it was, we changed 24 out of 92 of our top leaders [over the span of] two months. What I'm now telling my head of HR is I want to get ahead of this and figure out how to develop leaders because it's expensive to change a leader.
How do you institutionalize that How are you making those decisions about who can and can't change
We have a huge HR development program. We talk about how do we develop them, how do we develop the next-generation leaders, including the next CEO. How do you develop the next-generation operating committee leaders We literally review it with the committee. We've been doing this for 10 years and especially for the last three in detail. It isn't just about succession planning, it's how many people are we getting in which areas. It's how you address some of the issues on diversity. It's hard work. You've got to be willing to put people into roles sometimes before they're ready and see how they do. In fairness, I'd say probably people did that with me about half a dozen times during my career. I thought I was ready. In hindsight, I didn't know what I didn't know, which wasn't bad. You've got to institutionalize that deep in the DNA and as a company, we've lost very few of our leaders except when it was time for them to be changed. We also do it in a way that you don't see an article by you all or on the front page of The Wall Street Journal about leadership changes. We just do it very smoothly and we treat people with respect on it. I lost very few of my top leaders that I wanted to keep.
You say you're spending a lot of time watching startups. Where do you see Cisco's next-generation competition coming from
We called it four and a half years ago. My PR team kicked me when I said it at the Wells Fargo Conference, but I said it's going to be white label and it's going to be free software. I said it won't be IBM or HP servers or Dell. We'll beat them. I said it's going to be an architectural play for us to beat them but then it will be an architectural play that allows us to lead against white box or bare metal or free software. That's starting to occur now, too.
Sony [one of our customers] will show you how our total cost of ownership with ACI [Application Centric Infrastructure] is 40% less than a white box solution with free software. It's about cost of ownership. That's before you talk about architectures coming together and that's before you begin to combine architectures for business outcomes, but you can imagine somebody who does a white box solution who saves 10% on a switch and doesn't get the business outcome or worse, introduces a security violation that causes them $100 million to a billion dollars of damage. CIOs are not going to do that, especially if they understand it and the CEO understands it.
So that white box is really more of an initiative up at the very largest cloud providers
No, I think it's a fair challenge across the board.
This is even enterprise applicable
Yes, but the difference was if we'd waited and we'd stayed on boxes it would be a tough one. We've moved over the last five years -- and I wouldn't have said this if I didn't already have the plan in place -- to architectures. So our ability to beat the white box is actually pretty good.
John, do you plan to get stronger in storage
We have focused on storage through partners and that has served us very well. At the present time we're focused with very key strategic relationships with NetApp, EMC and IBM. We will introduce storage at the edge, which we've been very upfront with, for the [Unified Computing System] UCS-type capability, and we're very much aware of what's occurring in the container-type capability as well.
What about leapfrogging those partners and getting stronger into the flash storage market as that is really starting to take off
We do flash at the low end in terms of our acquisition there. It's more of a philosophy issue. There's probably, just for discussion, across our architectures let's assume that we're moving into 15 areas. Each one of them can do acquisitions or partnerships. We've got a lot of opportunities in software, which I want to move faster on and a lot of opportunities in security, which we want to move faster on, a lot of opportunities on consolidation, which I want to move faster on. The importance of focusing on where does storage fit into that and are you better off fulfilling it through a partnership role vs. doing more of it yourself is an important decision for us. We are making the decision at the present time to do it through partners.
I want to just get an update from you on some key market initiatives. Talk about the Intercloud effort. Where do you stand with that and how is it starting to pay off for enterprises
Nobody questions us anymore about our cloud strategy or the uniqueness of it. What people see is a simple concept, i.e. to do in clouds what we did in networks. Networks were really clouds to begin with. We were drawing clouds 20 years ago.
Before it was cool.
All of a sudden you had AppleTalk and TechNet and SNA, etc. and we said these are proprietary networks, i.e. proprietary clouds, and it's going to come together underneath a single IP infrastructure where it will be completely transparent what networks you run over. You will see the same thing if we do our job right with Intercloud. It literally will be the vehicle so that you won't have a Google cloud that's separate from the VMware cloud that's separate from a Facebook capability that's separate from Microsoft that's separate from a Verizon or a Deutsch Telekom or a Telstra cloud. What we're building is the ability to share that very smoothly with common policy, common application-centric infrastructure capability, common security and workload balancing. That's really powerful.
Then you can say what do we want to accomplish as a company and where do you run your workloads for whatever combination of reasons. That doesn't mean the IT department goes away. Actually it's the reverse. They've got to now be more effective than ever and they can't lock themselves into solutions that lose control over what I think will be the winner and loser philosophy about how quickly you digitize your company.
Talk about the software-defined networking market. If I understand it correctly from what I've read from both companies, you're about neck-in-neck with VMware in terms of announced customer installations of SDN deployments.
I don't buy that, however.
Where does it stand
It's very simple. To me, you don't have a separate network for a software-defined network and a separate network for your physical network. We learned that with ATM and Ethernet and others. Too expensive, you can't share the information. We're going to lead the SDN market but we're going to brace for what people are after: programmability, lower cost, faster speed. So regarding the announcements about large SDN implementations, what you want to ask about is not how many people did you give it away to for free or off a software license. If you watched when Sony presented last quarter, it already has 300 customers with ACI capability and many of them well into production. Ask VMware how many software-defined NSX implementations they have in volume.
They've claimed 400 customers.
If they've got 400, ask them to give you 20 names to go talk to and you pick and choose because if they've got 400, you should be able to get 20 easy.
OK, we'll do that. What are you seeing in terms of enterprise adoption of SDN at this point
I was at a company yesterday that was trying to share with me what customers want is SDN. I said, "I apologize, but what customers want is flexibility in cost and programmability." I don't know anybody who wakes up in the morning and says I want SDN. What they want is the benefits of programmability, investment protection, architecture, flexibility of open architecture and that's where we win. It's unfortunate that VMware cannily threw a hurdle into our EMC relationship because if you watch their economic payback from their acquisition, it's been a disaster, with Nicira, a disaster.
How so
Not producing any revenue. What did they pay, a billion
Yeah, and for a very early stage company.
For a very early stage company. Disaster.
The other thing I wanted to ask you about is Internet of Everything. When you look at it right now, what is the impact it's having on Cisco from a revenue and product transition perspective
Let's start at the specific number level that you're leading me to. It's about a $4.5 billion run rate and incremental pull-through growing at 40% to 50%, so it's a big number.
Is it project-based revenue
Usually these are whole architectural-based revenues and as long as you measure it equally, then you can see your trends. So it's both the core capability and the pull-through that we would not have gotten. Name the smart cities that aren't Cisco based. Pretty hard, because it's architectures. You have to know what to do with architectures on mobility or traffic or security or government services or lighting. These have to have a common approach and we have the products along with our partners to do what you need. I like where we are.
How long do you think it is before Internet of Everything is just an assumption, baked into everything we do
I think it's going to be in five years. The impact is every company, every country becomes digital. The Internet of Everything is really an architectural concept. People get it and it is about people, process, data and things coming together with a whole bunch of architectures underneath a cloud, security, mobility and everything else that goes with it. By 2020, almost every company will be well on its way to getting there and the majority of projects will fail. That's what a lot of people don't get because it's way beyond just connectivity of things. It is how do you get the right information at the right time and the right person and the right machine to make the right decision.
With all of that going on, to some extent you can despair for IT because there's so much to do and so much change, whether it's in your thinking or how you deal with the business. The business leaders in your company, the end users, what do they have to do to be successful What are the elements of successful IT people
I have some exciting news. IT is back in vogue. If you were to honestly ask CEOs four or five years ago how many of you even use IT to enable your strategy, much less really become dependent upon them, many executives considered IT just an expense item.
Can I push back on you a little bit People understand the CEOs, and that executive level, certainly understands the power of technology and the need to do everything you're saying. I'm not sure IT is in vogue because I think in a lot of cases, people are looking at IT and saying: "You haven't gotten us there and we really need to get there."
You're baiting me in a fun way, which I accept.
I'm just talking about what we see.
We're seeing it the same way. By IT being back in vogue, the CEO knows they have to become a digital company and a technology company. If you do that, you're going to have to have the capability to deliver on that. The question is do you get that in-house or externally It's still IT. But the key is whether IT can be fast.