Meg Whitman tells IT leaders everything they need to know about HP Enterprise

HPE

Meg Whitman doesn't shy away from a challenge. She led eBay from tiny startup to household name, ran for governor of California and, nearly five years ago, took the helm at Hewlett Packard and stabilized an organization stumbling badly from a variety of very public missteps. Having engineered the split of the Silicon Valley icon into consumer tech (HP, Inc.) and corporate-focused Hewlett Packard Enterprise, Whitman is now HPE's Chief Executive Officer.

The IT market is undergoing fundamental and rapid change owing to cloud, mobile and other powerful drivers. The competitive landscape in which this $50 billion startup plays is also shifting dramatically, with a slew of emerging players and the prospect of the largest-ever tech merger of Dell and EMC. No sweat, right?

In this installment of the IDG CEO Interview Series, we asked Whitman to talk directly to IT leaders about what the company split means for them as customers, and how HP Enterprise's new innovation agenda will help them transition to private and hybrid cloud. Whitman spoke with Chief Content Officer John Gallant about why HPE is better positioned than Dell/EMC to drive customer success, and about the company's strategy for hyper-converged infrastructure -- HPE's so-called Composable Infrastructure. She also shared insights on big data, cognitive computing, networking, high-performance computing and other critical growth areas for the company. Whitman discussed the challenges ahead and how she'd like perceptions of the company to change.

I really want to focus on what your strategy, as well as some of your recent moves, mean for our readers, who are senior IT leaders. If I'm a CIO or another top IT executive who has considered HP a strategic partner over the years, how exactly does this split benefit my organization?

First, I would say that the market is moving at lightning speed. I'm sure your audience says this to you every day. I've been in the IT industry for a long time. I've never seen it move this fast. And, in fact, part of the reason for the split was that we had to get smaller to go faster.

What they will find, I believe, is a stronger, more agile, more innovative company that is better positioned to help them transform their IT infrastructure. Virtually every customer I talk to has an aging, siloed, relatively high cost and not as flexible an infrastructure as they would like. And they're going to have to get to a new place with better security, using big data, enabling a mobile generation of users. Most people, not everyone, but most customers need help thinking through that and actually getting it done.

We hear terms like agile and nimble a lot. Can you give our readers some specific examples of what that means for them? Where is this agility actually showing up in terms of new investment or bringing things to market more quickly?

It turns out these two businesses -- both HP, Inc. as well as Hewlett-Packard Enterprise -- are two different businesses. Right? The PC and printer business is a scale [business]. We sell six PCs a second at HP, Inc.

The Hewlett-Packard Enterprise business is a solutions business. And we are now doubling down on innovation and R&D. We have introduced probably the fastest and most important innovation agenda that we have had -- at least since I've been here, for sure, and maybe in the last decade. A lot of long-time HP'ers tell me this is the best product lineup we've had in a decade. It is more focused.

[According to data provided by an HPE spokesman, R&D increased both in absolute terms and as a percentage of revenue over the past three years. R&D grew from $1.95 billion and 3.4% of revenue in fiscal year 2013 to $2.33 billion and 4.5% of revenue in FY2015.]

I'll just tell you personally, I went from running seven businesses to basically running four. And I can tell you I go deeper on each customer, I go deeper on the technology roadmap over the next three years and I work a lot more closely with our business units because I have more time. And that shows up, I believe, in results. We've strengthened our go-to-market and I'd be happy to go through some of the products that we've introduced in the last year if that would be of interest.

I definitely want to go into some of those key areas that you've mentioned. When you talk specifically about R&D and doubling down, can you put some numbers around that? What exactly would people see there?

Well, first of all, they would see research and development as an increasing percentage of sales over the last 4-1/2 years, even as we were cutting costs pretty dramatically. We expect that trend to continue with Hewlett-Packard Enterprise. When you break apart the two businesses, we actually spend a higher percentage of sales on R&D at Hewlett-Packard Enterprise than we did at HP, Inc. because, obviously, the PC business doesn't have a big R&D spend and it's a huge business. I think you will see quite a different picture of our R&D spend and we're going to continue to increase that spend.

When you talk about that innovation agenda, what tops that innovation agenda? What should people know about that?

Well, it really focuses around the four transformation areas that we outlined late last year. The strategy for Hewlett-Packard Enterprise is in many ways the same. We want to help people transition to what we call the new style of IT and we're organizing ourselves around four transformation areas that customers basically told us were their biggest pain points.

I'll recap briefly for you. How do we help IT leaders transform to a hybrid [cloud] infrastructure? How do you decide what apps you want locked down in your data center and touched only by your employees? What are you willing to have in an on-prem private cloud, a virtual private cloud and managed by the cloud and in the public cloud? And then how do you orchestrate all of that for the best asset utilization, the most flexibility and the lowest cost?

The second area, of course, is how do you secure your digital assets? We have both a services practice and a software practice there.

Then, how do you empower a data-driven organization? And we have a slightly different take on this because, remember, we have a very big high-performance compute business. We're practically the last man standing in high-performance computing -- only us, Cray and SGI. As the data multiplies, someone has to crunch that data and so we're really focused on the high-performance compute market. We have products like Vertica and IDOL that help gain insights from all that data.

And, finally, empowering a mobile and generationally different workforce. I don't know about your company, but our 50-somethings have a very different point of view on what they want at work than our 20-somethings. And that was of course why we bought Aruba, for great wired, wireless, LAN capability at the campus, branch and edge.

That investment really seems to be paying off.

It was such a great acquisition for us. It was really just like 3PAR, just like 3Com, it was complementary technology that leveraged our go-to-market, and frankly gave us legitimacy in the switch market. Obviously now everyone understands that we're committed to networking and a great alternative to Cisco in many, many instances.

Meg, I want to get into some of those specific product areas in a moment, but I want to explore more of the context here. Let's say I'm a customer that's committed to HP. If I'm concerned about the company no longer being the one-stop shop that it once was, or I'm worried about losing long-time contacts at the company or changes with my service and support, what are you telling customers like me?

Our TS [Technology Services] organization -- which is break/fix -- used to be one organization. But 4-1/2 years ago we actually split TS to support printing and PCs, and then the other half to support our server/storage/networking/cloud initiative. What we found was the skills to service printers and PCs are actually quite different than the skills to service Superdome Integrity X servers running SQL. So that has not been a big change for customers. If they bought our PCs and printers they got one set of TS execs and workforce, and if they bought our data center products they had another.

When we went through the separation, we actually had two separate contracts with customers because we had to invoice as two separate companies, they had to pay two separate companies. So to do that we went customer by customer -- by the way, value added reseller by value added reseller, of which we have 150,000 -- to get that technology change made there.

Then, by the way, most folks had at least two representatives calling on them. They had a data center person and a printer and PC business. Our global accounts had one person, and the global accounts will still run interference. If you're on the HPE side, they'll run interference for HP, Inc.

I have to say that customers have adapted to this beautifully. I think I've had one customer call me and say, "Hey, I wish I could still have my AGM [account general manager] in charge of all of HP." And I said, "Well great, you can. Don't worry about it, we'll do some one-offs for you."

But the truth is the people making the printer and PC decisions are almost entirely different than the executives that are making the data center decisions. So it's been a really seamless separation of these two companies. Customers got it. There's an expression in politics: When you're explaining, you're losing. And we didn't have to do much explaining on this, honestly.

A lot has happened at HP over the past decade. With all that was going on at the company, where do you think you've fallen behind? And on the flip side, where are you ahead of the competition?

It's hard for me to comment about where we missed the initiative prior to when I got here, and I didn't spend a lot of time trying to figure it out, because you can't change the past but you can change the future. So I wanted to make sure that particularly on the data center infrastructure side that we were best in class. We refreshed our entire server portfolio from top to bottom, whether it's rack or tower or the Gen9 ProLiant server. We've invested heavily, as I mentioned, in high-performance compute. We've got a refresh of our Superdome servers, we call them Superdome X and Superdome Integrity X servers.

HP

HP Enterprise CEO Meg Whitman

And we have very much focused on HP OneView, which is the system software that actually knits together many of our most important products.

If you think about converged [computing], I would say converged was in many respects our idea, and I think we didn't execute as well on converged as we should have. But you look at our new converged offering, HPE Synergy, and I think we have now caught up with and perhaps surpassed the competition. We're really excited about Hewlett-Packard Enterprise Synergy, which I'm sure you know is basically our Composable Infrastructure.

Another two-sides-of-the-coin question. What do you see as the biggest challenge for the new HP Enterprise right now, and what do you see as the biggest opportunity?

I'd say the biggest challenge is this market continues to move at lightning speed. There's a new competitor every day in some element of our business, and it's not usually the big guys. It's a Silicon Valley startup. We have to make sure that we are leading the market, not following the market. But if we need to follow, we need to follow fast, and then add that HP magic to our infrastructure products.

We're very pleased with our Helion Cloud that is getting real traction in the marketplace, over 200 new customers in Q1. Version 2.0 of Helion OpenStack is fabulous. In fact, I was just talking to our own internal IT team yesterday, who was super-excited about the next generation of Helion OpenStack.

So it is a brutally competitive market. I have to say this is far more competitive than my background at eBay. Everyone thinks that the consumer Internet space is competitive. This is a whole new dimension, and I think it's only getting more competitive as the underlying technology is changing. So that's, I think, the biggest challenge -- speed and agility.

And the biggest opportunity is we have a very well-capitalized company in Hewlett-Packard Enterprise. It is smaller, more nimble. I will put up our engineers against anyone in the industry for enterprise-grade technology. We've got a real opportunity to redefine Hewlett-Packard Enterprise for the next generation of enterprise infrastructure software and services.

What perception about HP would you most like to change now?

I think we've done a good job of changing the perception, but I still get asked questions. When I was in Davos, I did an interview and the fellow had actually not moved beyond a decade ago. I mean he was asking me about board drama, turnover of CEOs. I mean the old stories about HP, which by the way were not inaccurate, you know, five or 10 years ago.

We've actually had stability now for nearly five years. I think we've done a great job of revamping our go-to-market. The board is the best board, I have to say. We ended up fixing the board at HP before the split and then the split gave us an opportunity to get world-class board members. If you haven't looked at the board members for Inc. and for Enterprise, you should. I think both boards are the best boards in corporate America. Diverse, incredible skill sets that are relevant to the business and real team-oriented boards. So I think I'd probably love to make sure that people have moved beyond the chatter about HP five or 10 years ago.

In many interviews or stories about you, people will talk about the layoffs, the number of layoffs. What's the flip side of that; where are you hiring today? What customer-facing functions are being bolstered in particular?

In the last couple of years most of our reductions in force have been in Enterprise Services in complex European countries, which needed to happen. But as you know, it's very challenging to do that, so a lot of CEOs just don't. But we had to get the cost structure right for Enterprise Services. So a lot of engineering talent is being hired in offshore locations in SAP HANA, for example, our center of excellence for SAP HANA, for which we are I think among the best in the world. It's in the Philippines, in Manila. I mean if you ever have a chance to go to Manila and see our SAP HANA center of excellence, it's stunning. Big installations in Bangalore, Chennai [India], Sofia, Bulgaria, Costa Rica. So we're hiring a lot of very talented executives in these locations.

And you may say they're not customer facing. Actually they are customer facing in many ways because they're doing a lot of the work that the customers count on. We've invested in our go-to-market coverage model. A lot of the criticism of HP was we kept changing the people who were in front of customers, our AGMs or our AEs [account executives], which were the lead for both our services business and our infrastructure business. There's been a lot more stability now at the customers. We're doing a better job of covering the big companies. I remember when I went to the Research Board four years ago, three-quarters of the Research Board CIOs were like: "We haven't seen anyone from HP in three years." That has changed.

When you talk about improving go-to-market, what changes specifically for the customer?

So go-to-market is another word for sales effectiveness, and presales and the technology assistance that a customer gets. We have stability at these big customers, not everywhere, but most places. We also now have a CTO assigned to our biggest customers that can help them think through the underlying architecture that they want to deploy in their environments, whether it's on-prem as I said before or in a virtual or managed private cloud. What is the architectural roadmap that they want to deploy as they try to reduce costs and improve their ability to service business needs?

And then our presales organization, we have beefed that up as well. That is the arms and legs to can do a lot of the spadework with the CTO to help customers figure out what they'd like to do next.

Are there areas of hiring that you wanted to identify?

As I said, R&D, engineering. We have been increasing our hiring, particularly in our engineering functions. Anything that is R&D-oriented and sales-oriented, we tend to invest in and we've tried to reduce and be more efficient in the non-customer facing functions -- HR, finance, legal.

Although we have made a substantial investment in our own internal IT. This was one of those funny situations where the cobbler's kids have no shoes. We had quite some work to do in investing in our own IT so we could be easier for customers and partners to do business with.

I agree with you that this is a very exciting time in IT because the next-generation architecture is taking shape, although I'm not sure everyone is entirely clear what that will look like. Why is HP Enterprise a better partner going forward for customers than a Dell/EMC or a Cisco or anyone else?

As you well know, we obviously have a fantastic footprint across the globe. We can help customers implement anyplace, anytime. We talk about TS as a big important part of that, but first of all we can do anything anywhere. You want to set up, reinvent your edge IT system on an oil rig in the North Sea, we're there. You want to do it in Kazakhstan, we are there. There are only a handful of customers now that can do that effectively globally with their own people. So that's the first thing.

Remember I talked to you about our innovation engine and how we are really focused on innovation at the company? It's interesting because Dell/EMC, they have taken an entirely different strategy than we have. We decided to get smaller, they decided to get bigger. We decided to de-lever the company. We've got about $10.5 billion of cash on the operating company at Hewlett-Packard Enterprise. They have chosen to lever up. We have chosen to lean into new technology, like 3PAR all flash [storage], our next generation of servers, high-performance compute, hyper-converged Composable Infrastructure and our Helion OpenStack cloud offerings. I think they are about to double down on old technology and run a cost takeout play.

The way you should think about it, in my view, is Dell/EMC is now owned by private equity. What does private equity do? They rip out cost. And some of that cost is probably good to rip out and some of it is going to cut into, I'm certain, their innovation engine.

Two entirely different strategies. As one of my investors said the other day: "One of you will be right."

Clearly, there'll be a period of transition as those companies go through a lot of changes if that deal goes through. But why do you think they're banking on the old? What indication do you have of that?

Well, think about what's going to happen here. They are doing a private equity play. When you get bigger, when the market consolidates, what are you actually trying to do? You're trying to rip out cost and you're trying to rip out duplicate cost as well as cost costs. Just think about the R&D spend there.

The cost of the Dell/EMC debt is going to be much higher than they had originally thought because the debt markets have moved. It'll be probably in the $2.7 to $5 billion annual range. The entire budget for R&D at EMC is $2.7 billion. So there's going to be some impact in my view on R&D.

And then there's going to be, to your point, a lot of figuring out who's on first at what customers, go-to-market rationalization. Talk about losing your favorite sales rep. By definition, half the people are going to lose their favorite sales rep. I think from a financial architecture perspective, this could be quite powerful in the long run, but I think it's going to be a challenge.

We've got to stay focused on innovation, on customer needs, on presales, on architects, on these four transformation areas, while Dell/EMC is restructuring themselves. I think there's at least a two- to three-year opportunity where we have a chance to really show our world-class innovation and the ability to help customers migrate to a better place for them.

So let's talk about that opportunity. How will you take advantage of that from a strategic perspective? I've seen your advertising saying don't let your business be put on pause by Dell/EMC. What are you telling those customers and how will you capture the hearts and minds of those Dell/EMC customers?

There's really a two-prong approach there. One is to make sure that all the Dell/EMC VARs know that they are most welcome at Hewlett-Packard, so we are reaching out to Dell/EMC VARs who are nervous, because VARs make their living with these companies. I know this personally, because when I got here all of our VARs were nervous. So I recognize nervous VARs.

We're talking about our roadmap for innovation, our ability to sell with them, our ability to hunt new accounts for them. And this was the play we ran in many ways when IBM sold their server business to Lenovo. We're actually doing that exact same play, which was very successful for us.

And then for direct customers, we really talk to them a lot about what our capabilities are from a service perspective, from a software perspective and from an infrastructure perspective. One of the interesting things about the Aruba acquisition was we are now introduced to a lot of installed base of other companies, whether it be Cisco or Dell, because Aruba had a very long-term partnership with Dell. When I was at the Aruba Atmosphere Event, which is their big customer event, I probably met with 10 to 15 customers, virtually all of which were Dell shops, because they had a long-term partnership with Dell. In a funny kind of way, we have a lovely introduction to the installed base of some of our competitors, and we have to earn our way in.

I always say: "Let's do one or two things really well for you. You'll get to know Hewlett Packard Enterprise, you'll get to know our capabilities. And you can decide what you'd like to do with us next."

If I'm a smart Dell/EMC customer, what should I be pressing HP Enterprise for now? What can I get out of you?

First of all, from a storage perspective, our all-flash portfolio represents an incredible growth engine for us and real differentiated functionality for our customers. We're the fastest growing vendor in the all-flash array [market]. We're growing three times the market rate. Our vision is to accelerate the journey to the all-flash data center and enable a single, unified data fabric that links servers, storage systems and integrated appliances. That's the real point of difference and people are pretty excited about it, and the numbers bear it out. No one grows three times the market if they don't have a differentiated offering.

The other thing that they should talk to us about is Composable Infrastructure, and this is what we call Synergy, and it really brings together compute, storage, networking fabric through a single interface that's powered [by] HP OneView. It composes physical and virtual resources into a configuration that is unique to an application.

So it's a great developer tool. You can compose your infrastructure, get it running in production, and then it automatically releases compute, storage and network capability back into the pool if that application no longer needs it. This is very forward-looking new stuff that results in lower cost and more agility for the developers.

I think our line of servers is the best in the world. You just look at our Gen9 server; we're sharing many more three-year product roadmaps with our customers -- and the server roadmap is second to none. It is remarkable what we are doing there. HP Apollo, our high-performance compute, is remarkable.

And then they should look to us for cloud. We are number one now in private cloud. Private cloud is the easy way to get started as you want to cloudify your environment, because private cloud is on-prem, it gives you a 10% to 30% cost advantage versus a traditional data center environment, but you have control. It's a great way for customers to begin this journey to cloud. And then if you think about our cloud system automation tools to orchestrate in a multi-cloud environment, people should look to us for that.

Here's one of the things that is different than what I thought it was going to be. I thought most companies would pick a cloud provider -- [Microsoft] Azure, [VMware] vCloud, Helion, whatever it happened to be -- and they would basically move whatever applications they were going to move to the cloud, they would move to that cloud provider.

That's not what we see out there. People are running multiple instances of cloud, whether it's AWS or Azure or Helion or vCloud, and so they've got a multi-cloud environment orchestration and automation challenge that I think we're uniquely suited to help them with.

Those would be top of my list of things.

I would probably also say Edgeline, which is our new line of servers that is designed to be at the edge. I think it's going to play a real role in IoT, because remember in IoT compute has to be at the edge.

It was interesting, I was in an autonomous driving car in Berlin about two months ago. Have you ever been in an autonomous car?

No. How fearful were you in that situation?

This was sort of frightening, honestly. We had a driver, but he was not driving. The car was driving itself. And it brought home this compute at the edge for me, because think about it. You're driving along in the car and the sensors in the car, 360 degrees, are inputting data instantaneously. Okay, so is that traffic sign a person or is it a fixed object? As the car is changing lanes, is the car in front speeding up or slowing down?

So the data input is huge, and it has to be processed in real time. If you open up the back of one of these driverless cars, basically what they have in the trunk is a baby data center that is processing all this data in real time.

It brought home to me that what we have to do on the edge is we've got to have little baby Composable Infrastructure effectively that can combine server, storage, networking in a converged infrastructure. I think that's a super-exciting opportunity for us and I think we're leading in that.

I mean these are little components. This Edgeline server, we should like take a picture of one and send it to you. I mean it would fit in the trunk of a car. So whether it's in an airplane or in a medical institution or whatever happens, to me these Edgeline servers, I think, are pretty cool.

We talked about Dell/EMC, but let's look at another competitor, Cisco, for this next generation of infrastructure. Clearly strong in networking, strong in at least a portion of the server market, they have partnerships in storage and could make a bigger play there. They have all the elements. How do you view them as a competitor in that emerging IT architecture?

Obviously their core strength is data centers, switch and networking. And they've done a remarkable job over many years. What we see is there is an opportunity for us in the campus, branch and edge, because most CIOs want an alternative to Cisco. They would like to test out our networking operations in what they perceive to be a lower-risk environment, which is their campus, branch and edge, which I totally get. Right? If your campus goes down, it's okay. If your data center goes down, it's not okay.

But they get to know our data center switching and they get to know our data center capability via the Aruba acquisition, and that is actually working really well. What I say to customers is, "Listen, you need to inject competition into that market, you need to see what the latest and greatest capability is there, because you do not want to have vendor lock-in as you try to reduce your costs and improve your agility over the next number of years."

So I think we've got a real opening there. We're really the second largest, now, networking company in the world, at the tipping point I think -- certainly at campus, branch and edge, and perhaps even in the data center. From a converged infrastructure perspective, I think our HPE Synergy will now launch us ahead of Cisco, and so we've got to go earn that business back.

We're the only company with server, storage and networking under one roof, so we don't have to rely on partnerships, although we do have reference architectures, as you know. So we feel pretty well-positioned there. But they're a very strong competitor and we've just got to keep innovating and keep doing things that make customers' lives cheaper, better and faster.

So I want to ask about private cloud in terms of how long is this a stopgap? Do you see this as an ongoing market or how long is it before people move significantly more of their applications and workload to the public cloud?

So this is the million-dollar question, right? It is a big disruption in the industry and the question, to your point, is: Which workloads move on what kind of timeframe and where do they move to? Do they move to the public cloud? Do they move to a virtual or a managed private cloud? Do they move to a private cloud on-prem? And the way you have to look at this, in my view, is application by application and workload by workload. Because there are some workloads that I think will move quite quickly all the way perhaps to a public cloud: Web applications -- customer-facing Web applications that are not mission-critical applications to an enterprise.

There are some applications that frankly will never move. There are applications that are running on mainframes, as you well know, and I don't think they're ever moving anywhere. I mean I'm not quite sure what's going to happen, but a lot of these core banking applications I don't think are ever moving because the cost to move them is too high. So I think it's a continuum.

And it also depends on the industry that you are in. If you are in a customer-facing industry that is lightly regulated, my view is you may move applications faster. If you're in a highly regulated industry, either from a compliance perspective or a data perspective, you will move more slowly. There's no question that this is a trend and we're trying to make sure that we can help customers do what is right for them and for their environment. Hence our partnership with Azure. Customers that want to move some applications to a public cloud, we're not in that business anymore and we have a deep and tight partnership with Microsoft and our view is that you should move to Azure.

You mentioned earlier the differences between employees of certain ages, but HP Enterprise, Dell/EMC, Cisco, don't you all have a millennial issue as well? Any emerging company or younger company, the model is going to be almost completely cloud and increasingly so over time.

Yes. So if you think about new companies that are starting, they're running their IT infrastructure completely differently, and so we have to figure out how we can remain relevant to these young companies. And I'll give you a couple of good examples there in a minute. But the vast majority of the total available market is in legacy infrastructure. I mean by definition, if you've been around more than five years, you have a legacy infrastructure. I think there is a tremendous opportunity for us to help companies with legacy infrastructures migrate to a better place for them, to be able to compete with these new younger companies that maybe have an entirely different infrastructure.

But it's interesting, we're now watching some of these young companies that started out completely in the cloud, and now you see the cost escalating dramatically. For example, Dropbox, you probably know, built their infrastructure on AWS, and now have moved to a more traditional environment. Not a traditional environment, but they moved to a private cloud, because the cost was high. I think there's going to be an interception here as some of these new companies get to be big companies, global companies, with all the data privacy issues and data storage issues in Europe.

I think you're going to see cloud adoption in Europe, public cloud adoption in Europe, be slower because of data privacy regulations around keeping the data in country. And by the way, we now have to certify that that data is in country.

The ability for an HP-scale company is I think still super-important here. But you raised a good point. The world is changing fast and we've got to make sure that we can help our customers change to the IT infrastructure that's going to be best for them.

HP

HP Enterprise CEO Meg Whitman

And I'll tell you a little story, which I think you probably know. About two years ago we started a program called Pathfinder, where we are taking advantage of our position in the middle of the greatest renaissance in B2B infrastructure and security and big data that's happening anywhere in the world. It's right here in Silicon Valley. We basically said, "For those four transformation areas, we've got to partner with some of these young companies and integrate them into our solution so we can offer the newest technology best-of-breed." We just led the financing round for Mesosphere, and so our job now is to help companies understand what DCOS {Mesosphere's Datacenter Operating System} represents and then integrate it into their environment as they transition to a hybrid world.

We are increasingly forming partnerships and making investments in these small companies that we then curate for enterprise scale. A lot of CIOs tell me, "This is all great. There's a new security company being born in Silicon Valley once a week. What am I supposed to do with all these companies? Am I supposed to integrate them all into my environment? How do I know they scale? How do I know if someone will be able to service them?"

So we're going to try to be a curator, if you will, because if we integrate a young company into our solutions, we have to be able to support them globally. We have to be able to break/fix globally. By the way, I think that will make us very relevant to customers. Someone asked me the other day, "But Meg, when you integrate these young companies into the go-to-market powerhouse that is Hewlett-Packard Enterprise, these companies could go from $4 million to $40 million to $150 million to a billion dollars relatively fast and you will only own a small percentage of the company." That is correct. And, by the way, if that's true, we will have increased our relevancy to the CIO and I'm sure they will buy a lot more of the more traditional Hewlett-Packard Enterprise offerings.

But we can't buy all these companies. First of all, there's too many of them. And in security, by adopting companies to integrate into our solution, if another one comes along that is better for our customers, we move to that one and we're not stuck having paid $200 or $300 million for a company. It's a different operating model for Hewlett-Packard Enterprise, it's a big cultural change for the company, because we are used to selling only what we own, for the most part.

In the past in talking to HP executives like Mark Hurd or Leo Apotheker, we often heard that HP needed to become more of a software company. What is the software strategy today? Help customers understand how that strategy has evolved and what you're going to focus on.

I would make a differentiation between what I would call application software and system software. As you well know, we have always been a huge player in system software because our infrastructure does not run without that software. We will continue to lead in the system software area. And then, my predecessors actually made a number of bets in big data, with products like Vertica, in orchestration and automation software like ITOM and ADM, which is actually quite close to our hardware offerings, as well as our employee productivity software and things like that.

So we like the application software business. We're going to continue to invest in that business. But when people say we're not a software company... Actually, we are. I think we have about $3.8 billion worth of software products, which would make us the fourth or fifth largest software company in the world. And then when you think about Aruba, Aruba is actually a software company as well. What runs these Aruba beacons is software. But I would say it's a bit more like system software per se than application software.

Another way to look at this is, I guess, is what won't you do in software?

So we won't do ERP. We're not going to go buy Salesforce or Workday or any of those kinds of things. I think we will stick to our orchestration, automation, application lifecycle business, our security business and our big data business, but we will not be in the ERP business per se.

I also wanted to ask you about an area that's really building up steam: Cognitive computing. IBM is certainly staking a strong claim to leadership there around Watson. What is HP's strategy when it comes to cognitive computing?

Cognitive computing for now lives in a couple of different areas. First is, you can think of IDOL and Vertica as cognitive computing. Vertica has in its customer list some of the most forward-looking companies. Facebook, Uber, Airbnb are all Vertica users and basically what they are doing is cognitive computing. They're ingesting machine data, they're ingesting user data to basically gain insight around what's happening to their market and what their customers are doing.

Then we have some more upstream research in cognitive computing that lives in Hewlett-Packard Labs. Cognitive computing is basically ingesting all kinds of different data and making sense out of it. We do a lot of that. Listen, they're {IBM} doing some very interesting things. They're buying healthcare companies and stuff, but listen, our Vertica platform is remarkable as is our Haven OnDemand platform. I put us up against Watson every day of the week here. We're in a lot of customers where actually from a Watson perspective it's not as far along in terms of real-world applications as you might imagine from the advertising.

I'm looking for the brief explanation of how you want people to think of HP Enterprise. eBay was the online auction company, no question about the mission, the role. If you were to put it in a couple words, HP is the what company?

It's the company you can count on to transition your IT environment to the new style of IT.

From an employee perspective, in terms of reengaging the employees at HP Enterprise, what have you done to get them to own innovation again or to spur morale?

First is very clear communication about what we're up to and who we're focused on. Remember our consumer business went with HP, Inc., so we are entirely an enterprise company. Second is sense of urgency. The market is very competitive and we cannot have people working at HP who are not on it, running to the fire and moving fast. I say to our organization every single day: The future belongs to the fast. If we are not fast and thorough and the reliable technology partner, we will not win. But if we are, we have a very good shot at winning here. So I think it's about speed and it's about winning.