Steve Ballmer: The Churchill Club

A Conversation with Steve Ballmer, CEO, Microsoft Corporation
The Churchill Club
Santa Clara, Calif.
Sept. 25, 2008

KAREN TUCKER: Tonight’s program is entitled, Microsoft and the New Software Economy, a Conversation with Steve Ballmer and Ann Winblad.

Our moderator, Ann Winblad, is a long-time member and guest speaker of the Churchill Club, she is a veteran of the software industry, having started her first company in 1975. She launched Hummer Winblad Venture Partners in 1989, the first venture capital company that was formed to fund exclusively software companies, and since then Hummer Winblad has launched over 100 companies.

Ann, welcome. (Applause.)

Microsoft CEO Steve Ballmer with moderator Ann Winblad of Hummer Winblad Venture Partners discussing Microsoft’s role in the new software economy at The Churchill Club in Santa Clara, Calif., on Sept. 25, 2008.

As CEO of the largest software company in the world, Steve Ballmer is an icon and veteran of the technology industry. He, too, is a veteran of the Churchill Club. He first appeared on the Churchill Stage in 2001, and then again in 2003, and 2006, and we are delighted and honored, Steve, to welcome you back. (Applause.)

ANN WINBLAD: Steve, let me give you my personal welcome back to the Churchill Club. I looked back at the history of podcasts, and we last hosted you in May of 2006 in a great conversation with Roger McNamee. I know we’ve both spent almost three decades in the software industry, but I know I’m hard pressed to recall a more dynamic time than the last two years.

So this week there are 40,000 techies swarming the streets of San Francisco for the Oracle Conference. Last week, there were 15,000 in Las Vegas for VMware’s conference. And in the background, the U.S. economy was perched on a meltdown. I don’t know what happened today, so it still may be perched on a meltdown.

So before we get started talking about the software and IT industry, can you comment on your thoughts on the health of the IT economy, both here in the U.S. and worldwide?

STEVE BALLMER: Yes. I think there are probably a couple of things. Our industry is not immune to what goes on in the global economy. There is no question about that. And yet as I travel and talk to primarily people in the industry, in the tech side, in the telecom side, I would say given the current circumstances, people still see a certain buoyancy in the market. Part of that comes from the fact that perhaps the U.S. is harder hit than some other parts of the world, the market is global; some of that probably comes from the fact that the consumer side of the market has been less hard hit so far than the business side. And part of it comes from the fact that nobody is quite sure how bad the fallout will be either from the credit crisis or from the improved situation, whatever that may look like, as a consequence of the bailout. We’re one week before the end of our quarter, so I have nothing all that interesting to say. But, I think it’s probably fair merely to reflect on the fact that, at least for now, people that I talk to in our business are relatively, I won’t say optimistic, but are feeling better than you would feel if all you did was watch CNBC all day. (Laughter.)

ANN WINBLAD: Well, fortunately we’re too busy in this industry to watch CNBC all day. So, Steve, when you were here two years ago, you said that there were four trend that were important to the IT industry, and you named them as digital lifestyle; Moore’s Law continuing to drive smaller, cheaper, faster; software and service; and last increased digitization of work and lifestyle. You did critique Roger that he was using paper, and unfortunately I still am as well. So, first, are these still the top trends that you see in the IT industry?

And, second, a lot has changed, and one of our big topics tonight is, how is Microsoft going to compete more effectively in this new software market, one that’s characterized by advertising supported software, software delivered as a service, and personal devices increasingly replacing personal computers?

STEVE BALLMER: Well, it’s one of these things where if you use the same buzz words four years in a row, they’re not very interesting, or three years in a row, I guess, going back to 2006, but at the end of the day the things I talk about today are about the same things, digital lifestyle, all information going digital, software plus services, the themes are absolutely the same, and to varying degrees, and various speeds, those things are happening.

Now, I’m excited, in the last two years, we’ve made two years of progress on almost everything that we’re working on in businesses that we’ve had kind of a leadership position on, we’ve made two years of progress. In businesses where we’re not the market leader. It’s easy to say, oh, it’s terrible, you’re not the market leader in everything. I say, well, thank goodness we’re not the market leader in everything, we have a chance to grow, and come on, and enhance our position as well. And that partly reflects our desire to move into new things. And as you point out, there’s a lot going on in the industry, it’s a dynamic place, and there are a lot of great opportunities for us and other companies.

ANN WINBLAD: So we talk about how the industry has changed, this question is asked to you every time you’re at the Churchill Club, and that is competition. Everybody rattles off the usual suspects, Adobe, Google, IBM, Oracle, Cisco, SAP, did I miss any companies here?

STEVE BALLMER: SAP is a partner, but keep going.

ANN WINBLAD: Okay. Great.

STEVE BALLMER: Especially since they’re sitting right here.

ANN WINBLAD: So the last time you were asked this question in this venue you replied, maybe just being nice to SAP again, that the biggest competitive challenge was none of these companies. You had lots of competitors everywhere, but it was the ability to either embrace or compete with alternative business models on their own. Is that still true, or do you want to go through some of those competitors?

STEVE BALLMER: I’d actually  at the end of the day we have incredible capability as a company developing software and commercializing it, in all traditional ways, enterprise software, 15 years ago remember people would have said we had no ability to commercialize enterprise software, royalty, packaged software. I would say actually we’re pretty sophisticated at advertising-based software.

People will be quick to point out we’re only number three in the world, but we are number three in the world. And if there’s advertising-funded models out there for software nobody other than Google, perhaps, is better equipped to get after those. I don’t like not being number one, but I don’t know what I would say to you if we were number four or number five. We sell over $3 billion a year of advertising. We’re one of the largest sellers of advertising of any company in the world, and yet we’ve got more room to grow and progress.

There’s ad-funded software, a new area for us, we’re building, we’re growing, we’re developing. There’s software embedded in devices, TVs and phones, we’re building, we’re growing, we’re developing. There’s subscription-funded stuff that will come out from the cloud. We have a lot more, we’re going to talk about that at our professional developer conference in about six weeks. But, we have an incredible pipeline of enterprise customers, moving to cloud-based solutions. So I’m feeling, actually, pretty optimistic, and yet every day we’ve got to keep pushing.

ANN WINBLAD: So let’s talk about some of these models in greater detail. Software as a service, last week when Fannie Mae and Freddie Mac were removed, actually they just kind of went away from the S&P 500, a fast rising company, a Bay Area company called Salesforce took Freddie Mac’s place on the S&P 500. In the 30 years that we’ve known the software industry less than 30 core software companies have reached $1 billion in annual revenues.

Salesforce is quickly approaching that number, and could be the 30th company on that list with $1 billion in subscription revenue, 38 percent growth.

(Almatur ?) here in the audience has emerged as a de facto analytic standard with a $363 million run rate in revenues, and 60 percent annual growth. If I tack on the next four, or three even software as a service companies, these companies alone are going to rack up $2.5 billion in annual revenues, and subscription revenue.

You call this software plus services, combining the power of the desktop and the service. Is this embracing software as a service, competing software as a service, are we talking about the same thing, or are these different business models all together?

STEVE BALLMER: There’s business model and technology model, I think when people think about subscription on something in which you have an ongoing operational responsibility in the cloud, I’ll call that a subscription business. When we say software plus services what we’re really trying to emphasize is a balanced model of computation, where some stuff may go on in the cloud, but that doesn’t mean that things don’t still go on on rich clients, be those clients phones, PCs, TVs.

If you take a look we’ve got this product we call Media Room, the set top box set of software, and head-end software. Is it a service that AT&T instances, or is it software plus service? You actually have rich, intelligent software running in the cable set top box, and you’ve got rich computation going on in the head-end. So from a technology and architecture point of view, software plus service actually relates to the kind of balanced model of computation. Most software plus services things for the enterprise I think are going to wind up being subscription funded, and many software plus services things for the consumer will be a mix of advertising, transaction and subscription.

ANN WINBLAD: So Microsoft over the last 30 years when you joined Microsoft the mission of the company was a computer on every desktop, and in every home. Is that still Microsoft’s mission statement today?

STEVE BALLMER: No, no. We’re working on a broader agenda than that, and I love that. It’s still my favorite mission statement of all time, because it was a mission, it was a vision, and it was a scorecard. Our people loved it, you could count up every day, how are we doing, how are we doing, how are we doing? But, in a sense when you look at the footprint of the things we work on today it’s almost like each business deserves to have its own equivalent. And in a sense maybe we should make that formal, I think our enterprise business has a mission, our advertising business has a mission, our phone and TV business, our devices business has a mission.

Windows, they still  we still talk about a computer on every desk and in every home running Microsoft software. So it kind of varies. Today we talk about enabling people in businesses throughout the world to realize their full potential. And that’s fluffier, it’s not certainly a scorecard, but I think it gets at maybe kind of the fundamental aspiration that brings a lot of people into the technology business.

We are, in our industry collectively, we get to change the world, we get to give people tools to allow them to be more productive, more creative, more effective, more in touch, more in contact. And whether it’s in a phone, or a PC, or a server, or in the cloud, I think that’s kind of the core value proposition of information technology.

ANN WINBLAD: So is there a mission statement like, patience and tenacity, or something like that, that overarches the whole thing?

STEVE BALLMER: I wouldn’t call that a mission statement, I would call that a way of life.

ANN WINBLAD: Okay. So great. So let’s talk about the scorecard. We were good at the patience part in venture capital, so let’s just talk about the scorecard a second. So Microsoft recently had their annual meeting to go over their June results. The numbers were pretty amazing, Steve. For Fiscal Year ended June 30th, $60 billion in revenues, 18 percent growth, one of the best years in the last half a decade. Last year’s results included an impressive 26 percent growth in earnings per share. These are huge numbers. I mean, I think the startups in the room are wondering what the zeros look like on these numbers.

So if we look at the segmentation of the business, which you report, the largest business was just shy of $19 billion in revenue. That’s your business segment, with just under $13 billion in operating income. Client had just under $17 billion in revenue, and $13 billion in operating income. So combined these two segments made up $35 billion in revenue, and $25 billion in operating income. Yet, some refer to these businesses as your cash cows. Now, I’m from Minnesota, so I know a lot about cows, so let’s park those for a while and talk about the other segments.

STEVE BALLMER: Just a small comment, I don’t know what a cash cow is, if a cash cow is something that’s growing 15 to 20 percent a year, that’s okay. That’s okay, anybody who wants to use the word, I’m glad to have that cow in our pasture.  

ANN WINBLAD: Yes, and you know what, cows have pretty long lives. Only one of your business segments lost money last year, and that was the online segment, which is a pretty small business segment for Microsoft, relative to those other two, which brings us to the advertising-supported model. When you were last at the Churchill Club you indicated that at the end of the day it’s going to be the ability to really create a mass marketplace that’s going to give value to buyers of ads, sellers of ads, and the consumer. You said two years ago that this would be a five-year task, actually a little over two years ago. Can you give us a progress report in more detail on how you’re scaling, or going to scale in search and advertising?

STEVE BALLMER: It’s a five-year task. It’s a long-term task, that’s really the right way to go into it. Look, we are a small  in search, let’s not talk about ads overall. In ads overall we’re a much larger player, we’ve got a big footprint with our portal activities, what we’re doing with MSN, Hotmail, Messenger, so let’s really focus in on the nub of the question, as opposed to everything else, is what about search.

Search we’ve got about nine percent share in the United States. The United States is the  one of the worst markets in the world for the market leader. Google actually has much higher shares in most of Western Europe, and China is an outlier, U.S. is low for Google, et cetera. But, when you’ve got 9 percent in any market or less, it’s going to take us a while.

We’ve got a lot we need to do. We have improved the core relevance of our search results in an amazing fashion. We started nowhere for years ago or so, and we now have a product that delivers very relevant, very good results. We’ve done a lot to differentiate in some of the areas like image search, news, maps, et cetera. The market leader still has the brand, of course, that means search, that’s a big, big deal, and it’s something we’ve got to get after. And while we’ve made a lot of progress we still have a lot of work we need to do to try to fundamentally reinvent the search experience, and the search business model.

You don’t really in our business brute force your way into any market. You only really make major strides when you also play the game of redefining the category, redefining the experience for the user, redefining the business model. And we’ve taken some steps in that direction, but a lot of our URGs so far have really been involved in what I’d say paying the ante, the jacks are better to open in terms of core relevance and operation capability, et cetera.

So it will be a long haul. I told our investors that at the investor meeting that you mentioned, and yet I think the opportunity is great, and frankly speaking, if anybody is going to provide any real competition in search, and search-based advertising to Google, I think Microsoft may be unique in our ability to provide some of that competition. We don’t do that philanthropically or altruistically, but we see a real opportunity not just in search, but really in the digitization of the advertising market. So we’re going to keep our heads down, we’re going to keep working, we’re going to keep inventing. We’re going to do what we need to do to be real players in that game, and it will take a while.

ANN WINBLAD: So just one quick question about that original five-year plan, did that include the acquisition of Yahoo, or not?

STEVE BALLMER: Well, if you want an honest answer, I can make a joke, but I’m trying to think of the timeframe. We made the bid that we made last 2008. We had discussions I think that wound up showing in the press, with Yahoo in late 2006 and 2007. So I suspect it was at least in my head at the time that I was the Churchill Club, but obviously no concrete action got taken at that time.

ANN WINBLAD: So in your head is there any other big, bold move like that?

STEVE BALLMER: If you want to have share in search, it turns out you’d better get share in search. Many people say is there a substitute, who don’t you, I don’t know, go buy Foobar, they’re a big Internet company. Being a big Internet company doesn’t make you a big search company. We are a big Internet company. We are the third biggest Internet company, but translating general success on the Internet into specific success in search means you’ve got to do the right job in search.

So I would say there are many interesting things out there, and many things for us to think about over time, nothing specific I choose to comment about today, but as it relates to really improving our position in search, the way to get there is to go do it on our own. There’s nobody else out there. I mean, sure, you can go buy Seznam if you want to buy search share in the Czech Republic. (Laughter.) No, they’re the leader. Come on, man, you’ve got to all get to the Czech Republic and study how to beat Google, it’s the best case study out there.

ANN WINBLAD: Come on down if you’re going to Czech Republic. Anyway, Steve, if we go back to big numbers again, you’ve also said that you believe technology innovation in search has slowed a bit, and that competing with Google interest h next fiscal year and beyond will involve billions of dollars. I checked, and Google spends about $2-1/2 billion in R&D annually, and $2-plus billion annually for capital expenditures. How many billions are we talking about here before someone wins?

STEVE BALLMER: Well, someone is winning. It just ain’t us, so to speak. So there is a winner out there. The guy who is not the winner is going to have to spend some money. I set the expectations with our shareholders that we’d be willing to lose between 5 and 10 percent of our total operating income for several years as part of an effort with real progress to gain share in search. So, that includes depreciation on the capital, it includes the R&D. Not all of Google’s R&D, of course, is spent in this area, but if you’re the number two guy and trying harder, you may not be able to match them dollar for dollar, but you’re at least going to have to ante up, so that kind of tells you how much. There’s an amount of money we’re going to have to spend on the R&D side, and on the CAPEX side just to be in the game.

ANN WINBLAD: So one last Google question from me, you may have others from the audience, so it may not be the last Google question. With overwhelming victory anywhere in the software segment come the SEC and EU. What advice would you give Google on that front? (Laughter.)

STEVE BALLMER: I would probably keep that advice to myself. (Laughter and applause.)

ANN WINBLAD: So let’s move to some of the other 

STEVE BALLMER: Way to go, Ann.

ANN WINBLAD: We’re going to leave search and advertising here.

STEVE BALLMER: She’ll leave the regulators.

ANN WINBLAD: Yes. Let’s talk about virtualization. Some have said that this is the biggest disruptive force in infrastructure software. So far we’ve seen virtualization as a terrific test and dev environment. Microsoft has been very active in announcements recently with the partnership with Citrix, new product rollouts, Live Migration, new Hyper-V, a lot of trash talk between you and VMware. What are your thoughts on where we are in the utilization of virtualization, and given that client desktops far exceed the number of servers out there, do you see desktop virtualization in the near future?

STEVE BALLMER: I think we’re very  numerically, virtualization is very early stage, less than five percent of all servers are virtual. Let’s just take servers for a minute. Less than five percent of all servers are virtualized. Why? Well, servers are managed in a variety of different ways. Frankly speaking, the virtualization software in the market has been extremely expensive. My opinion, the way you manage virtualization has deviated from the way you manage everything else in the data center. And so we see real opportunity to, I might say, democratize virtualization. More integrated management, lower price, high quality, and so far I’ve got to say, we’ve had just a tremendous reception to the work we’re doing with Hyper-V and our System Center Management software around it. It doesn’t mean that people don’t still like what they get from VMware, but I see real opportunity to bring virtualization to a much higher percentage of the world’s servers than we see today.

In terms of desktop virtualization, there are many forms of desktop virtualization. You could say, hey, look, what’s really going to happen is, you’re just going to have a virtual image that executes up in the server, and paints to a dumb client. You can say, we’re going to actually put a virtual machine down no the client to provide better application compatibility as we move forward. You can say we’re going to do application level virtualization, and then stream down the virtualized application to the client. You can do virtualization of the desktops at the presentation layer. Again, there’s many ways to sort of apply virtualization techniques to improve the experience on the client.

Does any of this to me mean that in the long run a significant percentage of the kind of computing people do on clients today will move to the server? I don’t think so. It doesn’t mean it won’t happen. It doesn’t mean some of it will be important. We’re certainly investing. But perhaps I think the most important thing is the notion of really using hypervisors on the desktop as a technique to improve the desktop experience, as opposed to let’s just move all computation and recentralize it. I don’t really think in the world where people kind of like their own personal devices, people fall in love with their phones, their PCs, their laptops, I don’t think it all gets recentralized.

ANN WINBLAD: So, you’ve talked a bit about product virtualization, but I heard you mention the word “price” several times. Does that mean pricing pressure and total commoditization of this segment is one of your competitive strategies here?

STEVE BALLMER: Well, I think if you want virtualization on 80 percent of servers instead of 5 percent of servers, you’d better not charge three times as much as the price of the server for the virtualization. I mean, I’m not trying to be rude about it, for certain high end applications the approach that VMware uses is a perfectly good approach, but it’s not an approach that’s going to lead to virtualization of a high percentage of servers.

In every battle, it’s some combination of technology, price, value, all of those, integration, they’re all important techniques. I think the guy who is the market leader ought to recognizes that only 5 percent of the servers are virtualized, and you’ve seen already some moves in VMware to kind of respond to what we’re trying to do to go much more broad with virtualization.

ANN WINBLAD: Speaking of the new fearless leader of VMware, let’s move to cloud computing. So cloud computing has a front and center row in the software industry right now. If there is a buzz word, it’s cloud computing. VMware did become the most recent billion dollar revenue software company last year, and Paul Maritz of VMware has said the data center of the future will inevitably have federation between computing that happens in the data center, and computing that happens in the cloud. What is your definition of cloud computing, Microsoft’s vision of cloud computing, and how does this fit with software and services?

STEVE BALLMER: Well, I thought I knew what the word “cloud computing” meant until I sat with Ann and a bunch of venture capitalists this morning who used the word completely differently than I would have used it.

ANN WINBLAD: That was the other venture capitalists.

STEVE BALLMER: That was another venture capitalist. It is not all that complicated. Today you could say we have a form of cloud computing. We’ve had hosted servers now for a number of years. Why is hosted servers not called cloud computing, or has not historically been called cloud computing? Primarily because nobody went back and re-architected the underlying server software to design specifically for the kind of scaling, fault tolerance, geo-replication, security that you would want in an environment that’s multi-tenant and shared. So when people talk about cloud computing, they’re really talking about outside the firewall, and software that’s been specifically architected to be managed and propagated in a certain fashion.

So do I think there’s a relationship between the cloud and the data center? Absolutely. People don’t want to re-write applications depending upon where they happen to want to instance the software. So I think there  I would agree with Maritz on this notion of federation, and kind of a certain level of symmetry and homogeneity between the two environments. But I think when people talk about cloud computing, they’re talking just about taking some stuff, putting it outside the firewall, and perhaps putting it on servers that are also shared or storage systems, that are also shared perhaps with other companies that they know nothing about.

ANN WINBLAD: So the press has been talking about Project Red Dog for some time, and some have referred to it as the ECT for Windows. Can you tell us more about this Steve?

STEVE BALLMER: No. (Laughter.)

ANN WINBLAD: Will you be telling us more about this?

STEVE BALLMER: In six weeks. We’re going to talk a little bit about some of our stuff, well, we’re going to talk a lot about all of our stuff in the cloud computing, software plus services world at our Professional Developers Conference, which is six weeks from now.

ANN WINBLAD: So, Steve, you love developers, you say that over and over, you’ve danced for them. We won’t make you do that tonight.

STEVE BALLMER: Thank you. (Laughter.)

ANN WINBLAD: They gave me the taser, that was my moderation. So about a year ago Amazon Web Services had 160,000 registered developers, that number grew to 400k by June 30th. If you look at the open source area, just take a couple of open source companies, I see Marten Mickos here from MySQL, or we take a young company like Maven that builds platforms for Java. Maven alone has had three million developer downloads. Do developers still love Microsoft?

STEVE BALLMER: I think there are many developers in the world. We do very well on balance with the broad development population. And the question you have to ask is, how do we do with certain sub-segments. And I think we have opportunities. We do well on balance, and we have opportunities for improvement with certain classes of developers. If you take a look, and I’ll do this kind of, on the client we do very well. On the server, there’s two application types where I would love to see us improve. The first and perhaps the most significant is in the area of technical computing, scientific computing. You might say, oh, well, nobody cares about scientific computing. We care a lot about scientific computing. There are about five million scientists and engineers, they use a lot of compute power. One of the frontiers of computer science is taking sort of modern science, and doing simulation and modeling, big, big area. They’re probably not the developers you were talking about.

The other area for us is what we would call the Web workload. About 40 percent of Web servers run on Windows, and about 60 percent of Web servers, roughly, run on Linux, and typically Linux means PHP, and blah, blah, blah, blah, our stuff means .NET, blah, blah, blah. How are we doing, 40 is less than 60, I don’t like 40 less than 60, I don’t even like 60 if it was better than 40. Big numbers are good in the game of share, and we’ve got some work to do. And there’s no question we have work to do. And, actually, I think some of the stuff we’re going to talk about as we roll out so-called Project Red Dog and others really should start changing the game to some degree, because I think at the end of the day cloud computing, as much as anything, will be dictated by kind of the interest and the degree to which we capture the imagination of developers.

ANN WINBLAD: So developers out there expect a lot more love from Steve.

So, you’ve also stated that there are three big opportunities for growth through innovation at Microsoft. We’ve talked about a couple of them, software plus service, combining the power of the desktop and server, expanding the presence on the desktop and server, and with developers including new tools like Silverlight. You’ve also said that consumer technology, including defining a new era of consumer electronics is one of these innovation opportunities for you. We’ve not talked about the new era in consumer electronics or mobile devices. Microsoft, like you like to say, through innovation, patience, and tenacity, the way of life, hit a home run with Xbox. There’s no question about that. Certainly Apple, Google, Nokia, a little company in Canada called RIM, have a nice position, starting position, in the mobile market. Can you talk about your strategy for the future of Microsoft with the consumer and consumer devices, and specifically with smart phones?

STEVE BALLMER: Well, consumer devices, I think you really have to think phone and TV. Now, it may be a game console attached to TV, it may be a set-top box. But at the end of the day, there’s a big screen in the living room, the little screen you carry with you, and the screen in which you’re going to be up-close, personal, and detailed. That’s kind of where the PC fits in.

In the TV case, I’ll just sort of talk about that briefly before I turn to phone, I think we’ve got so much good stuff going between our Media Room, or the work we’ve done with Media Center and extending it out through the TV, the work we’ve done with Xbox. And yet, if you really say today, is it absolutely obvious what will be the device that brings intelligence to the big screens in homes, I’m not sure that that’s an answered question, even though I like the kind of technologies and assets that we have better.

The smart phone, it’s kind of interesting you talked about the start of the other guys. Which of the following companies sold the most phones last year, Nokia  smart phones, Nokia, Microsoft, Apple, RIM or Google? Who is number one, smart phones? Nokia. Who is number two? Microsoft. Who is number three? RIM. Who is number four? Apple. Who is number five? The guy who sold none until last week. (Laughter.) I say things that are mathematical, and you guys take them the wrong way.

ANN WINBLAD: We remember, it’s all scorecards, Steve. It’s scorecards.

STEVE BALLMER: It’s a game we’ve been in now four or five years. We have learned a lot, we’ve done a lot of good work. We’re on kind of version three or so of our phone. Our phones are pretty good. We don’t have all of the things  we’re playing a strategy, we’re trying to have software separate from hardware. It’s easier to play a strategy where you have hardware and software closely coupled the way Apple does and the way RIM does. We sell more devices. And yet the market presence of the devices is certainly much less, feels much less than the market presence of Blackberry or iPhone. Google is trying to play the game more similarly to the way we do. Nokia is an interesting juxtaposition where they’re trying to have proprietary hardware/software, but they’re trying to take the same software and open source it. We’ll see where that really goes.

Mostly what you have is a market where  and there are still a lot of smart phone stacks that are kind of randomly developed. I mean, LG has a stack of its own, Samsung has a stack of its own, in addition to the work that the rest of us do. I think the real question is, if you go out five or ten years, this year there are about 125 million smart phones sold. I think if you go out five years, it’s going to be a billion smart phones. Ten years, someplace in there, it will be a billion smart phones sold. If it’s a billion smart phones sold, will it be primarily guys who build proprietary hardware/software stacks, or will it be primarily a world in which you have software and hardware innovation that proceeds separately?

I think if those of us who are kind of that mind-set can ensure that you can get the software without compromising the hardware choice, software and hardware’s independent points of innovation will trump the guys who are more end-to-end proprietary. That means, in a sense, I think, the Linux mobile guys, us, Symbian, maybe Google with Android as they get into the game, we’re kind of battling for the big part of the market. That doesn’t mean Apple and RIM won’t make a lot of money and have great success, but they’re probably restricted in some senses to a certain maximum. Even Nokia, the biggest guy in the phone market, only has, what, 33-34, something like that, percent of the market. And if you want to reach more people than that, you sort of have to separate the hardware and the software issue.

We, in a sense, have the most mature of those platforms, ironically, of Linux mobile, Symbian, Android and Windows Mobile, we have the most mature of the platforms, we’ve made the most progress. We’ve got over 65 vendors building hardware with Windows Mobile. We’ve got a lot of work to do. We’ve got marketing work. We’ve got branding work. The smart phone market is a little bit like the PC market in 1983, kind of when the BIOS was first being introduced as an architectural layer for hardware/software separation. So we’ve got our work cut out for us, but I’m actually optimistic that the approach we’re taking is kind of the right approach for us. And I’m optimistic that kind of the current position we have leading in that segment we can extend.

ANN WINBLAD: So I’ve got three more quick questions for you, Steve. We’re going to leave technology for a bit, but do a little bit more counting. Your R&D expenses in 2008 were about $8.2 billion, 14 percent of revenue, pretty consistent number as a percentage of revenue. Recently your CFO indicated that you would spend more on acquisitions, that you might spend more on acquisitions than research for the first time this year. Is that true, and how would you characterize the pace of Microsoft’s acquisitions?

STEVE BALLMER: I think our acquisition is a pretty consistent pace. We’ve been buying about 20 companies a year. It turns out if you buy a few big ones, you can get to $9 billion pretty quickly. The way you’re doing the counting sounds a little bit like the bailout, though. You’re kind of mixing balance sheet expense versus operating income, $9 billion is on the balance sheet in one case, and $8.2 is on the income statement, $8.2 net present value every year is a lot more than $9 billion. So, yes, we’re still essentially a company that does primarily our own R&D, but we’ll buy about 20 companies, most of them we’ll probably pay any price from $10-15-20 million on up to a few hundred million. We’ll do a few acquisitions from time to time that are a little bit bigger than that. This year it was interesting because we did both aQuantive and FAST, which were both acquisitions over a billion dollars, and we closed the transaction to buy Danger, maybe even in the same period we closed the transaction to buy TellMe, which were both over half a billion dollars. So it was a particularly active year for slightly larger transactions financially. But we’ll keep buying about probably around 20 companies a year.

ANN WINBLAD: So while you’re in Silicon Valley, a number of these acquisitions, quite a few were out of the country, a lot out of Silicon Valley. How has globalization of your markets affected your acquisition strategy, and how relevant is Silicon Valley?

STEVE BALLMER: There’s a lot of great work that gets done in Silicon Valley, but not all the great work gets done in Silicon Valley. I mean, I know that’s a little bit of a tough message for this crowd. (Laughter.) But if you look, certainly the percentage 

ANN WINBLAD: There are some people from out of town here.

STEVE BALLMER: Okay, fine. The percentage of acquisitions we do, if you had looked at it, I would say six-seven years ago, a very high percentage of everything we looked at buying was actually in the Valley. Today, it is a high percentage, but not a very high percentage, and certainly there’s a lot of great companies out there that start and we’ve probably bought 10 companies in Israel, we’ve probably bought four or five companies in France, we’ve bought a number of companies in the Eastern part of the United States. There’s just a lot of great people doing great work across the United States and across the world.

ANN WINBLAD: So at the end of the fiscal year, you had 91,000 employees, but one of your employees today told me there’s 100,000 people, so you’ve added another 9,000 since June. I don’t know if that’s right number of not.

STEVE BALLMER: I hope that’s not right. I would go check it on my laptop, but I’m not sure.

ANN WINBLAD: I can tell you what group they seem to be in.

STEVE BALLMER: Okay, good.

ANN WINBLAD: So, Bill has retired, we all see him a lot more these days with his new acting career. (Laughter.)

STEVE BALLMER: He’s hot, baby, he’s hot.

ANN WINBLAD: We can’t wait to see series part three of the Seinfeld commercials.

STEVE BALLMER: You may have to wait a while, but go ahead.


STEVE BALLMER: That was a two-week campaign, but man, did it get people talking for more than two weeks. Anyway, I’m not letting you finish.

ANN WINBLAD: So his hands are off the rudder, he’s got an office somewhere else, as I was driving here I heard him speaking to the U.N. today. He’s working on poverty, while you’re working on innovation. How hard is it to steer a ship of that size, and how challenging is it to balance your incumbent businesses with these  keeping Windows and Office relevant if you have great success in these other businesses? Is there a point where the balance is really, really hard to manage?

STEVE BALLMER: Well, I would say, look, running a diverse company, the thing that’s interesting today at Microsoft versus, say, seven or eight years ago is, we are more diverse in our business footprint, in what we’re trying to do, and actually in our revenue stream. We have real businesses today in advertising, in Xbox and hardware. And I loved operating a business that was all in one business. It’s kind of fun, you come to work you have your hands on the rudder every day. As the business gets more diverse you really are depending upon a strong leadership team to steer the ship.

Yes, Bill decided to retire, to go part-time, and that’s a decision that he could make, because of the strength of the leadership team that we have in place, not just at the level of my direct reports, but several layers down in the organization. And the only way to run a company, you say how can you do it, if I really had to feel like I was running Windows, and Office, and servers, and Xbox, and mobile, and advertising, we’d have a real problem. That’s not the way it works.

And in some senses each of our diverse businesses gets to build  sure, it’s all Microsoft. But, there’s the diversity of approach there’s a diversity of culture, there’s a diversity of business model. And I don’t mean in kind of a national sense, or an ethnic sense, I really mean the land of Windows feels different than the land of search, feels different than the land of Xbox. And you’ve got to promote, and accept some of that diversity. You’ve got to have leaders who drive it, and yet we’ve got to have a set of shared goals and shared values across the company at the same time.

ANN WINBLAD: So you joined Microsoft 28 years ago, it was a relatively small company, and the Churchill Club is about learning. Clearly, you have personally driven a lot of Microsoft’s success. So what would you tell the CEOs running smaller companies today are some of the key learnings, or maybe the top learning that you would tell them about the next 28 years of their career?

STEVE BALLMER: This may sound sort of trite, but it’s really true. Hard work is important, energy is important, leadership is important, accountability is important, hiring good people is important. But, there’s one thing that people tend to not talk as much about that is also really important, picking the right businesses to be in is really important. (Laughter.) No, I don’t mean  I don’t really mean that in a funny way. What’s the expression, the important drowns out  the urgent drowns out the really important.

Sometimes it’s easy to think so hard about what you need to do to kind of paddle the ship, you stop thinking about, am I really investing in building, doing the right thing. He most important decisions, the most important thing that happened at Microsoft, in terms of our historical success, is sort of three things. Number one, Bill decided that microprocessors were a new opportunity, software and hardware are different businesses. And if that hadn’t been right, all the hard work in the world wouldn’t have fixed it.

Paul Allen decided we should be in the applications business, if we hadn’t, we’d be in a whole different place, and basically you could say some combination of Bill, and, me, and Dave Cutler decided we should be in the enterprise business. And now we’ve made a set of similar decisions about Internet, this, that, the way we go after it. But, the decisions on what to invest in, maybe I’m speaking to the choir here, since you’re in the venture capital business, but what to invest in, and what to do I think people tend to under-emphasize those as sort of key aspects of learning to be a good business person.

ANN WINBLAD: So we’re going to go to some questions from the audience, but I have one last question for you. I know you have season tickets for the Sonics, why did you let them leave Seattle? If you had bought them, my partner John Hummer would have retired from the VC business and come and run them for you. So he’s extremely disappointed.

STEVE BALLMER: If you had told me, baby, we’d still have a team. It’s a whole big mess. Who wants to be the guy who buys a basketball team that has no arena. Howard Schultz, who runs Starbucks, did that. We tried to get kind of a new arena built, a lot of people in our community tried. I guess it’s time to buy season tickets for the University of Washington Huskies and watch them beat up on Stanford every now and then. (Laughter.)

All right. I thought I’d throw out a little raw meat before we go to Q&A.

ANN WINBLAD: So we’ve covered the economy, we’ve covered a broad space of technology, and we now just covered all of sports, and our major university. So before we go to Q&A, let’s give Steve a preliminary thank you for sharing his thoughts with us. (Applause.)