Steve Ballmer: Stanford University Graduate School of Business

Transcript of Remarks by Steve Ballmer, CEO, Microsoft Corporation
Stanford University, Graduate School of Business
Stanford, California
March 15, 2007

ROBERT JOSS [Dean, Stanford Graduate School of Business]: Steve, I think you can tell people are glad to see you back.

STEVE BALLMER: Very friendly, thanks.

ROBERT JOSS: Well, let me just start with kind of an opening question, and I’ll ask a few, and then we’ll actually turn to our students for the real hardballs.

STEVE BALLMER: OK.



Microsoft CEO Steve Ballmer (left) answers a business student’s question at Stanford University. At right, Robert Joss, dean of the Stanford Graduate School of Business. Stanford, California, March 15, 2007.

ROBERT JOSS: But let me just ask you to say a little bit about Microsoft. You’ve reached a new plateau now. You’ve launched [Windows] Vista and a new version of Office. And, of course, you’ve got some interesting challenges, you’ve got Linux in the enterprise, you’ve got Apple and Google in the consumer space. So, maybe the best opener would be just take a few minutes and tell us how you see the company and the state of play, and some of the opportunities and challenges right now.

STEVE BALLMER: You know, one of the questions I get asked a lot — and it sounds like kind of a silly question in a sense, and a very important question in others — what is Microsoft? Are we a software company, are we just a Windows company, what are we trying to do, what are our goals, what are our aspirations, what do we want to be when we grow up, in a sense?

And our core competence is writing software. That really is the thing that we know how to do. We write software for very kind of broad horizontal things, but then how to apply that, how to think about it is very important.

Now, if you ask, what are the things that I think characterize our company and sort of really talk to the issues you’re talking about, number one, we’re a fundamental believer that the power of software is going to continue to positively transform society.

We just did something where we wrote down 70 technologies that are going to change the world over the next five or six years. Bill Gates and a team of our technical guys wrote them down, and actually when they first came back with them, I said, you’re not allowed to have 70 of them. I’m kind of a more simpleminded guy; give me five. And Bill said, no, we spend $7 billion on R&D; I want to at least have $100 million on each, I get 70. (Laughter.) And I remember that from math camp, so to speak. (Laughter.) That worked.

And yet when you stop and think about it, the future of the way people interact kind of with their television, the future of reading, how many people read on paper, how many people don’t, why, what does the technology have to do to change and morph; the future of the way software programs get developed. We just bought a company down here that’s in the speech-recognition area, the future of speech as a technology; there’s so much innovation to get excited about. That’s kind of who we are.

And the good news is there’s plenty of opportunity from a business perspective if we just kind of follow our noses on the kind of interesting innovations that are out there.

In a business sense we’re actually now at a point where I kind of think of us as having to have four different business muscles, all around this technology. We grew up as kind of a PC desktop company. Never was quite sure what anybody meant by it, but people always told us, “You’re a desktop company.” Over the last 17, 18 years, we’ve tried to be an enterprise company. And people think of the IBMs and the Oracles of the world.

Now we’re trying to build two new muscles. We’re trying to be what I’d call a modern consumer electronics company, a company that does consumer devices where software is the hero, with hardware interwoven, as opposed to the old style consumer electronics where hardware was the hero, and you might put a little software in it.

And we’re also focused on being an Internet, information, and advertising company.

And you used the word “plateau.” I sort of don’t see a plateau. We finally reached a milestone; took us a little longer than we’d like to ship this Windows Vista product, but it’s in the market. And sort of in a sense we’ve enabled for ourselves the next generation of growth. In some cases it’s very obvious how we pursue, our market continues to grow, good for the PC industry. We’re moving into new areas where we have new competitors in the enterprise space. You mentioned a couple of them in your question.

In the consumer electronics area, Xbox is taking off, Internet TV is taking off, both good for us; challenges in some of the other competitors. I always love to remind our shareholders that areas where we don’t have share we can probably build more market cap than areas where we already have high share. So, in a sense I see kind of a positive in all of these areas.

I got asked the other day by an investor, and maybe I’ll kind of summarize the answer to your question, “Are you a growth company”, which there’s probably folks who have invested money professionally in the audience — I never have. So, I had to ask, “What is a growth company?” What’s a growth company? I don’t know. Is a growth company a company that’s investing in the future? Then I said we’re a growth company.

And so I said, “What do you call a growth company?” She says, “Well, you’ve got to grow 13 percent a year.” And I said, “What do we have to grow 13 percent a year?” Growing 13 percent a year is trivial if you mean revenue, do we have to grow revenue 13 percent a year? She says, “No, no, I guess I didn’t really mean that.” I said, “All we’ll do is we’ll just sell more hardware. We won’t make any more money, but we can grow our revenue 13 percent.” She said, “No, no, no, not revenue.” And I said, “Well, then what?” And she said — I said, “Do you want operating income to grow 13 percent a year, or EPS?” It’s actually easier to grow EPS than it is operating income, because you get to buy back your stock and do all that kind of stuff. (Laughter.) All that financial stuff, mostly second year stuff. (Laughter, applause.)

And she finally said, “No, you’ve got to grow your operating income 13 percent a year.” And I said, okay, well, at least now I know what the test is. I’m not going to tell you whether I’m going to make it or not, forward-looking statements, don’t make those, blah, blah, blah. (Laughter.)

But then I said, let’s just say 13 percent for three years, we make about $20 billion of operating income, so that means we’d have to grow operating income by about $10 billion in the next three years. I don’t know how many people in the room — try to name a company in the world that makes $10 billion of operating income that’s not in oil and is not in financial services. You will get some, but you’ll be hard pressed to get there very quickly. And I said, “So you want us to grow a whole 10 billion,” and I said, “I can’t make that commitment, but I can say there’s plenty of opportunity in what we do.”

And so we’re not really at a new plateau, I think we’re at a new kind of jumping-off spot.

ROBERT JOSS: So, interestingly, some of these new growth markets do not have their foundation in the Windows platform, which has been your historic source of growth and success. So, what have you learned about competing in markets where Windows isn’t your fundamental de facto platform to start with?

STEVE BALLMER: In a way what I would say is our core platform is the technology base we provide. And let me just take Xbox. I know, is Windows not part of the gaming market? I think that would be consistent with your question. And yet we use exactly the same graphics system essentially for people to write videogames that we use that allows Toyota to design and model next generation automobiles. So, we’re not using necessarily the same brand and same image, but we do get a chance to reuse technology.

And in a sense if you’re in a company that’s really an R&D driven company, once you’ve spent an R&D dollar, you’ve got to sell the R&D to everybody. Especially in our business, there’s just no marginal cost.

So, if we spend dollars on graphics systems, we ought to sell it to people in PC desktops, we ought to sell it to scientific rendering farms, we ought to sell it to people in videogame machines, we ought to sell it everyplace that it’s popular. And that’s why it’s really driven us to work on these four business models.

You know, people said Windows was a boat anchor in a sense for us as we attack the enterprise, because we were just a little old desktop company. Well, today we’ve got a business that’s close to $10 billion in enterprise software, and Windows and all those technologies were helpful. And I think the same thing can be true in the online world, I think the same thing can be true in this kind of modern consumer electronics business.

ROBERT JOSS: So, moving a little bit from strategy to organization, I’ve read recently that you set up a system of three presidents of the major businesses, but I’ve also read that, Steve, you’re a really hands-on guy, you love to dig into the data and the numbers. So, what’s it like being one of your presidents, and how do you actually operate with these three folks, and still keep yourself informed?

STEVE BALLMER: You know, if you actually ask, is there enough to do, without micromanaging every detail, I guarantee we have no shortage of things that need to be done.

And one of the things I think people miss when they think about this is just how complicated any large organization is, the Stanford GSB, Microsoft, any financial and services institution.

You know, when I think about my agenda, I have to represent the company externally, that’s 35 percent of my time right there, whether it’s speeches like this, customer calls, people want to see the CEO. That’s a third of my time.

I have roughly a third of my time that is tied up really in what I would call coaching, mentoring people, formal management meetings, operating meetings where we do review strategy or people or execution in the businesses that report to me, board meetings, recruiting meetings. I’ll go out and try to help recruit people, one-on-ones. Let’s all call that time that I can’t micromanage people, but where I can help. (Laughter.)

And then about a third of the time I actually get to work on things which I want to focus on. And generally I will tend to focus on things which are big growth opportunities or which are tough strategic challenges. And in every one of our businesses we probably have one or two things that fall into that category, and I’ll spend most of my time with those leaders and with their teams in those areas, and that seems to be okay. But it is not like — I can honestly say I think I’m the only person who can say they’ve either been the number one or number two guy in companies from all sizes between 24 and 75,000. I don’t think you can find anybody else who would tell you that they’ve basically run companies of all those sizes.

And I will tell you there does come a point where — there’s a point where you think you can know everything that’s going on, and then there’s a point where you discover you stopped knowing everything that was going on really three years earlier, and you’ve got to evolve to a new approach. And we kind of hit that about 1998, 1999. In my case, I didn’t even pretend I knew everything. I thought I knew the business side, and I thought Bill knew the technical side. And probably around ’97 things really had already morphed, and we probably figured that out by about ’98 and ’99. And now we’ve sort of systematically been building around teams that really drive this stuff, and trying to add extra value as opposed to everything else.

ROBERT JOSS: So, listening to your breakup of time, an awful lot of it you might call leadership and people management. Could you perhaps share a little bit some of your own personal insights about leadership and managing people?

STEVE BALLMER: There are two or three things which I think are most important for leaders to do, and yet don’t always focus on them very well. Everybody likes to say leaders have to insist on accountability, and that’s right, but there’s a yin and a yang to that, too. You want to insist on accountability, but you also want to insist that people stretch themselves and push themselves.

If the game is I got done what I said I could do, and you’re really rigorous about that, you will encourage people to set their goals low. And so how do you manage the yin and the yang of stretching people and yet at the same time holding them accountable. That’s sort of a key leadership dimension.

How do you manage the yin and yang of wanting to sort of delegate, and yet knowing enough of the details that you’re an intelligent participant in these decisions, the funding, the key issues that your people are involved with? I can’t say I’m going to know nothing about most of our businesses, and yet I don’t want to micromanage our businesses. And how do you hit that blend?

Businesses are complicated, that is true, and yet leaders have to deal with the yin and the yang of the fact that things are complicated, yet one of the leader’s great value-adds is simplifying things.

The truth of the matter is the perspective that you get, the more you’re at the 50,000 foot level, to try to be able to say, here are the three or four things that really matter, the key principles, and yet not make them seem so mom and apple pie is to be meaningless. And a lot of the times we have problems with leaders, the leaders simply aren’t trying to simplify, they’re trying to show they understand the full complexity of the situation.

And leaders really do need to hit the exact right balance on what I’d call the optimism realism curve. If you’re not realistic, you lose respect from customers, partners, shareholders, press, employees, if you’re not realistic. But if a leader can’t be optimistic, if a leader can’t say, life is going to be better, we’re going to take share, we’re going to improve this situation, if you can’t sort of handle being optimistic and realistic at the same time, I think it’s tough to be a great leader.

So, it’s probably those four leadership perspectives that with my guys I wind up spending the most time on is getting the right balance for them, for me on those kind of four dimensions.

ROBERT JOSS: So, as you mentioned, you’ve been there now going from 24 people to 75,000 people. There was clearly a time when you were well-known for being the place to be, this is the place to join, innovative, young, entrepreneurial company. Now as you compete with people like Yahoo! and Google and Apple, certainly familiar names around here, how do you keep the place innovative, exciting at 75,000 versus when you were at 24?

STEVE BALLMER: I think there are a couple dimensions. One is, how do you continue to get the best and the brightest? The fact is we just finished our best recruiting year ever. And I’m very proud of that, excited about that, and yet there’s more recruiting competition than there has been in a long time. The economy is good, particularly here in the Valley things are good. Hedge funds want programmers these days; that’s kind of a pain in the neck for those of us who didn’t have the financial services industry impeding on our issues. So, getting the best people remains a big issue for us.

And you could say just as Windows deserved a competitor, and we got one in Linux, so did Microsoft for talent deserve a competitor, and we’ve got one at least sort of showcase competitor for talent. That’s ok. It’s kind of like Linux; people said it would wipe us out. Did it wipe us out? No. In fact, our market shares today are stronger than they were three or four years ago. And the same thing in the hunt for talent; you’ve got to put your head down, you’ve got to say we’re going to work at it. We’ve got a compelling proposition.

If you want to work on the broadest set of kind of interesting ways to transform the world through technology, there’s no better place to work than Microsoft. That’s a fact. If you want to engage in broad innovation, there really is no better place to be.

Now, some people actually think it’s innovative to work on the next generation of development tools, and some people don’t. And I’d pick something that sounds really techie and da-dah-dah, dah-dah, but there will be breakthroughs in the next 10 years in the way software gets created. And somebody is going to make a lot of money out of that, somebody is going to show a lot of innovation out of that. We’re about the only place on the planet that people are pushing that agenda as aggressively as I think it ought to be pushed.

The truth is a lot of people, particularly people coming right out of school, love consumer products, and you’ve got to be hot with the consumer products. Apple has done a nice job with iPod. We’ll see how they do with iPhone; I don’t know. We’ve done a nice job with Xbox. I’ll bet we have a few Xbox enthusiasts here in the audience. And our phone product is now starting to get some real traction. Google has obviously done a nice job with their search product, but frankly have had a hard time getting traction in some of the other areas. And so on the consumer side we’re all just out there kind of really working to earn our chops.

I will tell you one thing I have learned. All companies start out young and age. It turns out every year every company is one year older than it was the year before. (Laughter.)

ROBERT JOSS: That’s another quantitative skill. (Laughter.)

STEVE BALLMER: The truth is great companies have to figure out how to remain vibrant, despite the fact that they don’t remain essentially run necessarily by 24 and 25-year olds.

I always tell our people, what’s the brand we should aspire to that’s been around for a long time, and yet people think of it as kind of a leadership dynamic brand. And it’s hard to point to a lot of those, because there’s sort of a — I don’t know about natural destruction, but I guess I can point to two of my almost alma maters. The brand of Harvard, the brand of Stanford in education, those have been good and sort of dynamic brands, leadership brands for many, many years. And in some senses I think that’s what great companies have to aspire to is brands, and brands are made by the product and services that you offer, that you want to keep them great not just for 10 years, 15 years, 20 years, but really out into the future, and we’re going to try to stay as dynamic as any institution in the world.

DEAN JOSS: And some people would also say that brand is a product of your culture. We hear so much about company culture. What’s the culture of Microsoft like, in your view, and what do you do as the CEO to try to change it or nurture it or strengthen it, and what’s your role?

STEVE BALLMER: We have kind of two personal characteristics and six values that we think a lot about. For personal characteristics we like people who are very bright and very intense, very bright, very intense, very bright, very intense. (Laughter.)

DEAN JOSS: Unlike you. (Laughter.)

STEVE BALLMER: Unlike me.

And I think that’s part of the culture. And you get good and bad with that. You can prioritize — I mean, you might say, doesn’t every company prioritize those two things, and I think the answers is no. We do, and our culture reflects that. But we also have integrity as a core value for the company, and I think it’s just important you always have to remind yourself about that. Accountability is a core value of the company. We want people making big, bold bets. It’s ironic, people are saying, “Are you going to innovate?” Well, who else would try something as nutty as we did when we set out to do Xbox, a big, bold bet that not too many people would have made, and yet we made it.

And so there’s collaboration, dedication to making others better. That’s one where our culture needs ongoing nurturing and improvement and evolution. And we’ve done I think a really good job outside with our business partners with the complexity of our R&D process, and we also need to have that same kind of collaboration inside.

So, there are some things like that that I would say are hallmarks and characterize our place. The desire to change the world also means we want lots of customers. Winning to us is sort of measured in the number of customers we have versus the number of people who engage in any activity. There’s on the planet about 10 million software developers. I want them all using our tools. That’s kind of what we come to work — and any day we find someone who’s not, it’s interesting, yes, why not, why, why, what about this, what about this, what would it take, what do we have to take, is it price, is it features, is it capability, is it innovation? That’s a part of our culture.

DEAN JOSS: And what in particular do you do perhaps to keep strengthening that? Are there things that you do behaviorally or where you spend your time? And you were talking earlier about a third, a third, a third in that. Can you break that out at all in terms of reinforcing these things?

STEVE BALLMER: Basically I didn’t know this. I didn’t know this even when I became CEO — I’ll say that — which was seven years ago. Everything I do is a reinforcement or not of what we want to have happen culturally. You name an activity. I did three things so far today. I gave a speech and did Q&A with 200 venture capitalists down at our campus. I gave a speech to about 700 of our employees. I had a one-on-one meeting with you, and I’m here.

Every one of those things sends a message. We have Microsoft employees in the audience, we have recruits, we’ve got interns; everything here is communicating about us and either reinforcing or not reinforcing what we want to do culturally. When we meet with the VCs, everything there is going to reinforce what their portfolio companies have a chance to think about us.

So, I have to think everything we do is actually a teaching and learning exercise in what we believe in, what our culture is, what our values are, et cetera.

ROBERT JOSS: And could you comment a little bit; you said, “I didn’t know that even when I became CEO.” What is there about not being CEO or then becoming CEO and you see that, do you think?

STEVE BALLMER: I should have known exactly what the job of CEO was when I took it. (Laughter.) I mean, I worked with Bill for 20 years before he asked me to be CEO. And we were partners, we’re doing everything together. And yet two things substantively change when you become CEO. Number one, you really do have bottom line accountability, and you will feel it in different ways than when you’re not CEO, as you know. And number two, really the tone at the top in terms of culture and values and what’s important is non-delegable. You can coach other people, but the culture and transformation and standing up for it and believing it and teaching it, that starts with the CEO. It’s the one thing — you can delegate everything you want in finance, you can delegate everything you want to in marketing, you can even delegate many things in strategy. You cannot delegate culture.

ROBERT JOSS: Okay, enough of the easy questions. Let’s open it up, and who wants to go first? And we’ll do as many as we can.

QUESTION: Thanks a lot for coming. And I’m really curious about how you made the decision to drop out of Stanford and join Microsoft. (Laughter.) And especially what you saw in them, and how you were thinking about that decision, and how exactly it happened.

STEVE BALLMER: I was trying to decide actually what to do for my first year summer job. I was about — basically Bill and I stayed in touch after I got out of Harvard. I went and visited him. First time I’d ever been in Seattle was the summer before I started at the GSB, because Seattle is kind of off the end of the earth — at least it felt that way back then.

And so we stayed in touch. I was going through the first year. I love first year, it’s great, I was learning a lot of stuff; wonder what I would have learned that second year.

ROBERT JOSS: How to be a CEO, right? (Laughter.)

STEVE BALLMER: Yeah, maybe that’s when you learn that.

And I was trying to decide what to do for a summer job. And basically in my mind I had four good thoughts. One was to join a consulting firm, still a popular option. One was to join Morgan Stanley, still a good option I suppose. I was looking at Progressive Insurance, which was a relatively small insurer then, quite large now. But I’d gotten to know the president when he interviewed me as an undergraduate at Harvard, and they were offering me kind of an assistant to the president job, and that sounded kind of exciting to me. And then I was looking at Ford Motor Company, because I’m from Detroit, my dad was with Ford. The then CEO I think was a Stanford GSB alum, a guy named Don Peterson.

ROBERT JOSS: Don Peterson, yeah.

STEVE BALLMER: And so I’m trying to decide what to do for the summer, and normally I guess you don’t fly back and visit everybody for summer jobs, but for a variety of reasons I had a week scheduled on spring break, I was going to go see all these guys back in their home offices, so McKenzie I was going to go visit in Chicago, and Morgan Stanley, and I went to Boston to see Bain and BCG, I went home to see Ford. I wasn’t going back to Cleveland — no offense to Cleveland, but I just didn’t have that as a stopping point. (Laughter.)

And Bill gave me a call and said, “Hey, what are you doing?” I said, “Well, you know it’s a two-year program.” He says, “Yeah. It’s too bad, it’s too bad. Maybe you should come up here. Ah, no, you’d have to leave now.” He said, “Hey, too bad you don’t have a twin brother. We could kind of use somebody like you around here now. Yeah, yeah, too bad. Yeah, too bad.” (Laughter.)

I thought about it for a day and I said, well, Bill is the smartest guy I had ever met, and I always liked the idea of — I mean, I wanted to work for smart people, and I didn’t have that uniformly in my Procter & Gamble experience. (Laughter.) I had it, but not uniformly.

So, really smart, and I didn’t know much about the business. I had written a few programs in my life, but I knew they were the world leader at an emerging thing. That’s about really all I knew about the business. And Bill and I had talked about it, but I was no expert.

And so I said — I called him back and said, “Hey, why don’t I end my trip to see all these summer job companies, and I’ll come to Seattle and see you.” And so I did. I came in and I talked to Bill, I talked to Paul Allen, who was his partner. Bill had his mom and dad work me over, which was his standard recruiting trick in the old days. (Laughter.) His dad is like 6”7′ and quite an imposing figure.

Bill actually left me in Seattle by myself. He was going on vacation for the first time, so he kind of — “OK, see you, Steve, enjoy my house, try to lock the door when you leave.”

And I thought about it, and decided, yeah, Bill and I agreed we try it. If it didn’t work out, I could quit or he could fire me at the end of the summer, and I’d come back to Stanford. So, I went and talked to the folks in administration, and they said, “That’s OK, we’ll take you back. Are you sure this is a good decision? Are you sure this is a wise decision?”

I called my dad, who thought I was just about the craziest — well, my dad thought it was nuts — (laughter) — completely nuts. Microsoft, what’s that? What is it? What’s software? My mom said, “Why does anybody need a computer?” (Laughter.) This is 1980. And my dad — my dad was an immigrant, and he didn’t go to college, and so he thought it was sort of triple nutty. And he had worked at Ford where everybody had — my dad thought it was nutty when I went to Stanford, because most of the guys he knew at Ford had Harvard MBAs, and he had to check it out. And when he found out this guy Don Peterson was OK, Stanford is okay, blah, blah, blah. (Laughter.)

So, my dad was not — that was the toughest call I ever made in my life, but I went on up to Microsoft and after six weeks I decided I’d make a real mistake, and I should come back for the second year of business school. And I told Bill, and he said, “Steve, Steve, Steve.” (Laughter.) “Let’s you and me and my dad go out for dinner.” (Laughter.) So we did. And at dinner Bill said to me, “Look, you don’t get it. You may feel like you’ve just joined to become the bookkeeper of a 30-person company, but we’re going to put a computer on every desk and in every home.”

And I swear that recruiting pitch to keep me wound up being the mantra of the company basically for the next 17, 18 years. Three weeks later, I bought a house in Seattle, and I cherished my first year. It was a lot of fun. I actually did learn a lot that was pretty important. I go to all my reunions. People seem to forgive me not having finished the second year. (Laughter.) And if any of you find something that you have the same kind of passion and enthusiasm and excitement for, see Dean Joss about it. (Laughter, applause.)

ROBERT JOSS: It worked out pretty well. (Laughter.)

QUESTION: I have a question about (off mike). Following up to you worked from 24 to 75,000 in 20 years. And when you compare that to Google, with a company where they become a bit big, somebody told them, okay, let’s put (off mike) in charge, and they (off mike) basically attracting people from many places, including Microsoft. So, my question to you is what is your view on these kinds of competitors and the position of Google versus yours, and whether it is possible today (off mike)?

STEVE BALLMER: Well, there seem to be two notions in there, and maybe you had them woven together and I had a hard time. There’s entrepreneurship, whatever that means, and there’s growth and whatever that means.

Yes, we went from 24 to 75,000 people over 27 years, and if you did a compound growth rate and headcount, you’d get a relatively low number, it’s just a lot of years and it’s very consistent.

I think with the particular competitor you mentioned, they’re trying to double in a year. I think that’s insane, in my opinion. It doesn’t mean they won’t do it well.

But we’ve been digesting a certain percent growth over many years, and what that’s allowed us to do is to build up a base of capable people who can take on more capable people.

I don’t really know that anybody has proven that a random collection of people doing their own thing actually creates value. (Laughter.) No, no, look, I’m not — until the scenario plays out, let’s just say it hasn’t played out. Google built one very good business, okay, but they only have one thing that they do. Everything else is sort of cute, but there’s one thing — (laughter) — no, come on, we do a lot of cute things, too. I’ll tell you about our robotics effort, but it’s not paying for me to come talk to Stanford Business School. (Laughter.)

I like to refer to us as a two-trick pony. And two-trick ponies are rare in the history of business. Desktop was a trick and server was a trick. The third trick, we’re trying to do online; the fourth trick, we’re trying to do consumer electronics.

And you can fill things in around that, and we’ve filled in a lot of things around the desktop, Windows and the applications, Windows Server and the database. And Google is trying to fill some things in around their basic footprint. But there’s still really one business, and it’s the search and advertising business, and okay, great, and it’s grown at a certain speed. In fact, they’ve grown faster to the 10 billion mark than we got to the 10 billion mark, but that just says the industry was ready for that to happen and good things have happened.

So, entrepreneurship to me is more about your ability to fill in through your own smart efforts, and about going to do the second or third or fourth trick.

And so in some senses growth — theory: There’s really four stages to business. One is I invent something. It may or may never take off. Two is I get something to critical mass. Three is I milk it like heck when it’s at critical mass. Four, I create a culture in which I can do one again, and I’m also prepared to go out — I or anybody else, we all have to be prepared to go out and acquire and start the loop again.

Great companies wind up having to be all four. Most companies don’t ever get out of the “I have an idea” and get it to critical mass stage. Some people get to critical mass and are never able to really let me say milk the idea. Those typically tend to get sold off. And some people can’t regenerate themselves.

Entrepreneurship to me is as much about all four parts of the cycle, or the first and second and the fourth. Right now Google is in this part of the cycle where they’re milking effectively. It’s a fun part of business; hey, we’ve got something good going, ba-bump, ba-bump, you follow your customers, you do all that innovator dilemma stuff, you just keep following your customers, collecting money, la-la-la-la-la, it’s good. (Laughter.) Hey, 20 percent, 20 percent, we don’t have to do anything new, we’ve just got to do 20 percent more next year. (Laughter.) That was the ’90s for us, or I would say the ’80s and the ’90s.

And now we’re really in the phase where you’d have to say, look, we’ve got to loop again. And our server business was really like a whole separate process, and we’re going to do it again, and we’re going to do it again. And it’s not just a question of the technology innovation, it’s also the ability to embrace multiple business models.

You know, when people say, is this a growth company, they could be asking one of two questions. Does the model you have continue to grow over the next several years; or number two, do we believe that you’ll invent new models of growth? And great companies probably have to be able to do both very well.

When I think about General Electric, and the thing that I think people ought to think about is brilliant about General Electric, they do both. And there are very few — Procter & Gamble, they do both, and they’ve been able to do both over a consistent period of time. A lot of financial services companies, Goldman Sachs, for example, they do both.

And so the ability to do both of these things very well, they’re both aspects of entrepreneurship. And so some of it is about invention and some of it is about growth, but those are almost different things.

QUESTION: So, I just came from a class co-taught by Eric Schmidt at Google, but in actually that class there’s a theme that A’s hire A’s and B’s hire C’s, and also the importance of mentorship. And you being kind of the first business hire and continue to be at the top of the organization, I mean, who did you get your mentorship from, and in the areas that you need to develop, where did you look to, because you were actually at the top all from the beginning to the end?

STEVE BALLMER: To the degree I got, quote, mentorship, I would have gotten it from Bill. I mean, Bill was just an incredible — Bill Gates was just an incredible source of learning and education for me when I joined, not about leadership and business so much as about the technology. Bill is an incredible strategy thinker, incredible strategy, business strategy thinker, in addition to an innovator.

But I would say I got most of my mentorship by being able to study other companies. You know, our business is built in conjunction with partners and customers. I could tell you during the decade of the 1980s I probably understood what was going on at IBM as a company as well as anybody who worked there, what was working from a leadership perspective, what wasn’t working. Because I was basically probably 25, 30 percent of my time was with IBM. I learned a lot from IBM. I spent a lot of time with Intel.

You can learn a lot of good things in a way by studying and asking questions. And even to this day if I really want to engage that way, I’ll just ask other people how they do things. When Jeff Immelt and I get together, we’re always saying, how do you do this, what do you do about that.

Miles White was a guy who was ahead of me one year at the GSB. I ran into him last week, and he runs Abbott Labs. How do you do this? What do you do about that? You’re bored, you’re trying to add people to boards of directors. We’ve got the former CEO of BMW, former CEO of Merck. How do you handle these things?

So, everybody, it turns out, wants to help you grow, if you’re willing to engage with a lot of people. I didn’t have anybody I’d point to though as a single kind of a mentor, other than Bill Gates.

QUESTION: You talked about the tradeoff and how you approached it when you were leaving Stanford, but could you talk about the tradeoff you’ve faced along your career on a personal level, and what continues to be your greatest personal challenge today?

STEVE BALLMER: Time. I mean, there’s no doubt time is the greatest personal challenge. When I’d tell you what percentage of my time is in every area, it’s because I have a detailed spreadsheet. I allocate this is the amount of time I travel, this is the amount of time I’m home, this is the amount of time I do this, and it’s got about — probably about 35 rows in it, each with a very specific budget, a nights away budget that I kind of negotiate with myself, my kids and my wife every year, because we have three youngish kids. And I’ll tell you I work hard to be able to drop the kids off in the morning at school/the bus, and I’m home in time for dinner. My wife would say I’m home in time for dessert. (Laughter.) At least I’m cognizant about that, I would say.

But the number one tradeoff on a personal front, it’s not like I’m a person with a lot of hobbies. I run in the morning because I need to or I gain weight. (Laughter.) I won’t even call it a hobby. (Laughter.) I play golf because I like to have one thing that I do. And I’ve got my family and I’ve got my work, and that’s it for me, and that works for me, that works just fine. But then I just try to balance the time between the two.

QUESTION: Both you and Microsoft are going through a pretty unique phase right now with the transition of Bill Gates. And I was wondering if you could talk about that transition. What was the first thing that came to your mind when Bill told you, “Steve, I’m taking off”, and how have you sort of managed that whole process going forward?

STEVE BALLMER: I mean, in a way it’s been more gradual than that, because it goes back to ’97 I think when Bill first started talking about having me be president. And so we went through a thought process and a transition. And then Bill talked to me in late ’99 — no, he must have talked to me about ’98 about being president, and then ’99 about taking over as CEO. And that was a big transition for Bill and for me. And then a few years ago, three years ago now almost, he started talking to me about the notion of not being full time, goes to be part-time at Microsoft but not being full time, and that was a less — I mean, in a public sense it felt like it was a bigger deal, but I felt more prepared frankly for that transition than the transition to CEO a few years before.

We’re working on that in a few ways. Number one, Bill and I agreed explicitly he’s got to go from being leader to coach. In some places where he would have thought of himself as having to lead the activity, he’s the coach of the activity.

We’re trying to moderate his external engagements, so the partnerships that he was primary on we’re transitioning to other folks.

We’ve made sure we put the leadership structure in place two years before he took off so that folks who were picking up part of Bill’s mission have a chance to be coach and learn about that part of their work very explicitly. Ray Ozzie, for example, is our chief software architect.

I’m having a chance to go through that, because there are sort of strategy priorities that Bill is focused in on that I have — anything Bill is really focused in on I have never focused in on, and anything I’m really focused in on Bill is not. We view ourselves as sort of close enough as a partnership group.

One of the things I didn’t think about as explicitly as I probably should have is it’s nice to have a partner. Anybody running a large organization, it’s nice to have somebody who you can really talk to like a partner. Bill and I have had that for 27 years, and now I’m forging kind of a new set of partnerships, and that’s been probably the biggest transition for me.

You know, with Bill it’s easy: You cannot replace Bill Gates. You can run a great company, but you shouldn’t think you’re going to replace a guy who’s a founder, a guy who knows where most of the stones are unturned, a guy who’s as smart a guy as I’ve ever met in my life, a guy who’s as famous as any guy I’ve ever met in my life. (Laughter.) No, I mean, seriously, you can’t say let’s plug compatibly, OK, do we have another Bill Gates? Sure, run him into the job. You can’t think about it that way; it’s wrong-minded.

There was a book we had to read when I was in college, sections of a book by this guy Max Faber about the routinization of charisma, and it talked about transitions from, quote, charismatic leaders and how those go. I went out and I dusted the book back off, I reread it, which was actually quite helpful to think about. And you see a lot of great institutions that have managed to routinize after charismatic leaders. I mean, China kind of went through this from Mao Zedong through to the present day leadership. You can have great things happen after great leaders, but you’ve got to think about it and be explicit about it.

QUESTION: (Off mike). How do you think about developing those masses of employees? (Off mike).

STEVE BALLMER: I think you have to say there’s probably two kinds of developments that people go through. There are people who are really functional in their approach. We’re trying to develop people as salespeople, as marketing people, as finance people. When you get to a company that’s got 75,000 people in it, at least a reasonable percentage of the people never want to be, never will be general managers. And then there’s a set of people who really do want to be, and you’ve got to make sure that you give the right kind of development opportunity to each one of them.

You cannot run a very large company that’s got operational processes without having some people who may not have the same general management aspirations that others do. I’d say we have a higher percentage of what I’d call high aspiration employees than almost any 75,000 person company in the world. On the other hand, it’s still not 100 percent of our employee base.

So, we’ve worked really hard to give people both functional development, as well as what I would call more general — either general management or strategy management. We have a bunch of engineers who never actually want to be general managers, but actually do want to be the strategy drivers for businesses. And we need to give them kind of the best of — it’s not technology development, but they need to get some strategy and general management development, even though they don’t ever want to really manage people. Some of our most influential people manage no one.

And so we really have the three models that we work on and think about in terms of development. Everybody who joins out of business school basically wants to be a general manager of some form, or a senior manager if not a general manager. That would be almost uniformly true. But if you join us as a sales rep in Philadelphia, your grand goal in life might be to be a better paid sales rep in Philadelphia, and we want to make sure we nurture that person in their career desire as much as we do the newly minted Stanford MBA.

QUESTION: So, as a leader, what keeps you up at night?

STEVE BALLMER: I sleep very well. (Laughter.) No, I mean, I actually do sleep very well. You know, it gets back to the issue I was talking about in terms of realism and optimism. You have to be realistic. So, it doesn’t mean everything is always going well or that there’s no threat, no clouds on the horizon. But I do have a fundamental optimism about areas where we’re weak we’re going to get strong, areas where we’re strong we’re going to get stronger, people we want to get we’re going to get. I do have a fundamental optimism about that.

I’d say the things that eat at me the most tend to fall into two categories. Number one is new business models. Learning a new business model or developing a new business model is so hard. So, what fall in that category? We got a new competitor about six, seven years ago called Open Source software, Linux particularly; different, radically different business model. I mean, people could say was Open Source innovative. The number one innovation in Open Source is it’s a very different business model. And if somebody came to you in your jobs after you got out of business school and say, hey, you’ve got a new competitor that has no price and has no cost structure, you might stay up a night or two on that one, you really might. (Laughter.)

And so we grew up with a business model. We had to learn to embrace this enterprise model. We had to learn — we couldn’t embrace Open Source as a business model. How do you compete with it?

We’re learning to embrace advertising as a radically different business model, it really is, because you pay me for what I do for you. And so none of the connection points that exist in other kinds of businesses exist in an advertising funded business the way that happens to work, so we’ve really wrapped our minds around that; so, business models.

The second thing which I also can dwell on is when things aren’t going well on the people front, whatever that means. It might mean somebody who works for me is upset with me. I mean, you’ve got to do what you’ve got to do, but I really don’t like having people upset with me. At the end of the day I worry about that. When we can’t get two key people to really collaborate in the right way, when we’re trying to recruit somebody and we don’t get them, or when we lose somebody. We don’t lose many of our folks, but when we lose somebody who I don’t want to lose, that kind of can get to me.

So, people related issues, and tough business model challenges don’t keep me up at night, but they’d be that kind of thing.

QUESTION: So, in the last few years, both India and China have (off mike) more specifically of open software. Do you see those developments as a threat or (off mike)?

STEVE BALLMER: Well, I mean, ironically there hasn’t been really a software — standard software products or service company other than perhaps (Waway ?) that really has emerged yet out of China and India. It’s a question of time, in my opinion. But there are some natural advantages to companies that are based in developed markets. It turns out if you’re based in a developed market, you actually have customers who have more money to buy from you, and you get more market feedback earlier in the innovation cycle.

So, even companies that are starting in many other places in the world tend to have a dual headquarters of some form in the U.S. It’s interesting because people want to make sure that they’re in the feedback loop that comes from this let’s say well finance customers here in the largest developed market.

Sixty percent of the world’s computer science graduates come out of China and India now. So, it is clear that there’s a big opportunity for us in the pursuit of the best talent to recruit more people from those two countries, whether it’s to recruit those people to move someplace outside their home country, or to recruit them to live in their home country. With the current issues, which I notice we have a number of people here who have spoken with non-U.S. accents — with the current issues with H1-B visas, in a sense there is even more pressure on any company that’s trying to get the world’s best and brightest to actually home people in multiple parts of the world. And if you care about computer science, that really gets you looking very strongly to India and then China.

QUESTION: A few months ago, Microsoft went through a big organizational change, and as a result Steven Sinofsky, who ran Office, is now over at Windows. And Steven Sinofsky is well-known for running a tight ship. He ships on time, and he’s pretty hands-on. On the other hand, some people are concerned that innovation might be stifled. So, I guess my question is, what do you think the key changes for Steven Sinofsky are being in Windows right now, and in general how does the organizational structure support innovation in the Windows division in particular?

STEVE BALLMER: Yeah, let me kind of give it a general flavor. The leaders do have different skills. One question any business has to decide is, are we going to try to homogenize and have one way we approach all problems. And I’m not actually making value judgments; I think it should be different for different companies. I don’t think there’s as much variation in the way Wal-Mart store managers run Wal-Mart stores as there is variation in the way Microsoft R&D organizations run R&D, and I think that’s to be expected in a certain way.

And we have managers who are much more systematic. Can they drive innovation? Yes, they can. The person the lady references is a guy who’s a very systematic thinker, he’s an engineer by training, he’s taught some classes actually at Harvard Business School, that’s what he did on his sabbatical, very systematic thinker.

And people think being systematic doesn’t mean being innovative, and yet some of the most innovative work we’ve done is the new user interface in our Office product, which, quote, his systematic approach generated. Our newest billion-dollar product is a thing called SharePoint. It’s phenomenal, it’s a piece of innovation that came out of the systematic approach.

There’s also advantages to the non-systematic approach to innovation, which has more been a hallmark of our Windows group.

And all of these things can produce. The key is to — at least in my opinion in our company the key isn’t to quibble with style that somebody uses or the approach, the key is to grade the results. And you just have to say, are these leaders, are these teams, are these groups that produce products that just aren’t innovative, but are innovative and popular, and we’ve had leaders who were able to do that fairly successfully with radically different styles across our company.

ROBERT JOSS: One more question, right here.

QUESTION: To the extent you mentioned following innovation into the market, I would love to hear your views on an R&D project which led to probably number one or number two market share, outside Xbox.

STEVE BALLMER: Sure. Let me talk about — I’ll talk about one that I think is going to be huge where shareholders could argue we were investing too early, but right now I’m very pleased by that investment, and that’s the area of IPTV or TV that’s going to get delivered over TCP/IP Internet style network.

We started working on TV over Internet basically in about 1993, 1994. We were sure it was going to get super red hot super quickly. We actually acquired a little company at the time down here for almost half a billion bucks. I think it was almost half a billion. Yeah, I think that’s right. It was either 180 or 440, and I can’t remember right now. (Laughter.) I hope it was 180. (Laughter.) A company called WebTV, and that’s actually part of the anchor pinnings of our Mountain View campus. And we’ve just stayed at this notion.

You know, TV over IP connections is really about two things. One, it’s about an improved, richer, interactive TV experience. And two, it is now becoming the enabler that is allowing the telephone companies basically to enter the TV business, in addition to the cable companies.

And so we’ve been at this thing, what, 13 years, and in some senses just now we’re really starting to see people, the telecom companies really get after this as a serious proposition. And so whether it’s AT&T or Verizon who we’re working with in the U.S., Deutsche Telekom, but this is something where basically we had nascent R&D, we did an acquisition 10 years ago, so I consider that now all nascent R&D basically. We rolled it all together, we stayed persistent, we stayed patient. Everybody else who was around the area went out of business, went to do something else. And today I’d actually tell you I think our customers are going to take a big share of the TV business, and the innovations will all be innovation that builds on the innovation in our IPTV platform.

So, it’s another example where — that’s an example again where can I argue we were the first ones? There were probably some other guys in ’94 also, but we stayed patient and persistent.

The SharePoint product that I mentioned, SharePoint was sort of a concept for a more structured form of Web site that our guys invented. And today, if you go to business customers, it’s the hottest concept we have out there with our Office 2007 — I was just at a conference, we had about 8,000 business partners and customers of our ERP and CRM products together in San Diego. Everybody is talking about this SharePoint product because of the innovation that it brings, and it’s a category essentially that we created from scratch. It’s not a consumer category, so it gets a little less attention in the popular press, but, man, that’s been a successful innovation and a successful march to market. So, just a couple of examples.

ROBERT JOSS: Steve, I wish we had more time, but we don’t. Let me say how wonderful it is to have you back. You’re welcome back any time. And I know everybody here joins me in thanking you so much for sharing your time.

STEVE BALLMER: Thank you all. (Applause.)