NEWSWEEK: YOU were friends before you went into business together…
BALLMER: We met sophomore year in college. We lived down the hall from each other in the Currier House at Harvard. I had heard about Bill from some of the other guys, because we were both into math and science-type stuff.
GATES: I was completely opposite of Steve in that! didn’t believe in going to classes at all. There was a group who would sort of waste time, sit in my room, talk about things… But there were a couple of people who hung out in that group who’d gotten to know Steve and kept saying this guy is so dynamic, so energetic, you’d really like him.
BALLMER: The first thing I remember was going to see “A Clockwork Orange.” GATES: And “Singin’ in the Rain,” on a double bill. We took this thing called EC 2010, which was really cool. But you and I had talked a lot by then about our different ambitions. At that point, Steve thought maybe contributing to the government in a big position would be a good thing.
BALLMER: I was chiding you about going back to Seattle. “Come on, you got to get in a big pond like New York!”– I had that view of the world. When Bill started Microsoft with Patti [Allen], I wondered why you were doing so much administrivia.
GATES: What were you doing up at Currier, because most of the people there were anti the normal university?
BALLMER: Actually, my roommate and I got into Leverett House. Then we said, there’s more women up at the Quad, let’s try to transfer. Of course, it made no difference, it was mostly for…
GATES: Geeks.
BALLMER: Geeks, basically, like us.
GATES: There were two advantages. The male/female ratio was 1:1, versus about 5:1 elsewhere. And any meal, lunch or dinner, you could order a hamburger.
BALLMER: That was important to you.
So how did you come to hire Steve?
GATES: Microsoft started to grow, and there was more customer demand than we knew what to do with, and we ended up overcommitting ourselves. So I really felt like I needed a business partner to help me figure out what we should do. I even said to my parents: “The ideal would be if I could hire Steve.”
BALLMER: I was trying to decide what to do for a summer job at business school, and you called while I was trying to decide whether to do consulting or investment banking. I came here on Friday, and then you left on vacation. I took Bill to the airport, and then the next morning, I spent more time one on one with Paul, Which was super-weird. Because Paul wasn’t quite sure what to say to me.
GATES: Yeah, I hadn’t done a good job of briefing Paul. Paul and I owned the whole company. It was 64-36. But I thought it was so important to get Steve, I offered him a chance to own part of the company.
BALLMER: At first I was going to get 10 percent of the increase in profits over the base year. Then it was a 5 to 10 percent share and we ratcheted down. But Paul had never heard that.
GATES: Another funny part of this was, I go off on this vacation and I have to negotiate the salary with Steve. So we were on this ship-to-shore radio phone, where I kept saying, “Doo-Wah Doo-Wah! Over!” I think we started at $40,000. Everybody else on my vacation had been drinking and they kept saying, “Pay him whatever it takes! Get somebody to help you out! You look like you’re really stressed out!” So I agreed to pay $50,000. Which, of course, tums out to be like 10 to the minus fifth portion of your compensation … So Steve came in and said, “You hired me to get this thing straightened out. I figured it out: you need 50 to 60 more people.”
BALLMER: On a base of 30, Bill said, “You’re going to bankrupt this company! I did not ask you come here to bankrupt this company! You are supposed to make my life more sane!” The first year, we had some pretty knockdown, drag-out [fights]. But they were effective in helping us learn how to work together. I think we both wound up with a view that we’re going to have to invest long term, but also keep a fairly tight eye to control.
When you look back at the way the company grew in the ‘8Os, how much of it was design and how much was luck?
GATES: The fact that it’s natural for there to be a focused software company, as the microprocessor comes along, and that we’re there, and we see that, and other people really don’t see that–there’s certainly an element of luck in that. I had grown up playing around with software. And so the timing was perfect. As companies came along to compete with us, they made a lot of mistakes. WordStar. VisiCorp. Later, Ashton-Tate, Borland, Lotus. People who missed graphical interface, even though they were being told by us to go after graphical interface.
BALLMER: Getting IBM as a first customer for MS-DOS was more luck. It was a door opener. And we just started hiring. We didn’t know exactly what we were going to do with everybody, but we’d say, “OK, this year we’re going to go get 20 grads.” Patti and I were big stumpers out on these college campuses. And we decided to get rid of the agent structure in Europe, and build subs[idiaries]. Have them stumping around trying to hire people. “Hi, we’re little guys you’ve never heard of, but we’d like to hire somebody cool to run Microsoft Europe, or Microsoft Italy.” We were just these chuckleheads coming to town.
GATES: You’ve got to believe it’s going to be a huge business to build that foundation.
BALLMER: Then there’s the fact that since the time I’ve been here, Microsoft has basically had no cash-flow issues. It let us take the long-term view. Worrying about payroll is a very bad thing. When I got here, [Bill] used to worry a lot about it. He’d write down the number of employees, their salaries, the customers and contracts, and the cash coming in.
GATES: I kept balancing it…
BALLMER: To try to make sure that the thing wasn’t going to go bankrupt.
The basic business model of low cost, high volume–when did that crystallize?
GATES: We were very interested in being the guys to capture the volume from the beginning. We’d sit around and talk about “Will there ever be a billion-dollar software company?” Because the unit volumes that you have to multiply to are just incredible. We’d say, “Yeah, we think we can do it. We think maybe it’s possible.”
Has that ever steered you wrong?
GATES: The real stress it created was that I kept saying we should do a whole bunch of products, and we had limited capacity to do those products. So which ones do you decide to do, and do you try and overload things a little bit? When we did that original IBM contract, we committed a massive amount of software that was unwritten…
BALLMER: We tried to do too much, that’s one mistake. Because you figure if you’re not the first guy into a volume category, you’re really in trouble. And I think we wound up messing up networking. Networking wasn’t a volume business, although it is now.
What about the notion of the customer feedback loop? Steve, you’ve been quoted as saying your motto is “Do it, try it, fix It.”
BALLMER: Well, we never put out products that we thought were bad. On the other hand, unlike other people, we had enough courage to not have to think we had the perfect painting before we shipped.
Is that why you started hearing that Microsoft puts out a shoddy first version, then it improves it?
GATES: We never put out anything shoddy.
BALLMER: Let’s give a couple of concrete examples. Take our Outlook [information management] product. We knew when we shipped it there were things that if we took another year, or six months, we’d do differently. It turns out we found some bugs that we hadn’t found in the beta cycle. We’ll fix those very quickly. We’ll ship a small update, then a bigger update. Did we make a mistake? No, customers are benefiting. And we’re learning.
GATES: To be honest, I like the idea that people know when Microsoft introduces a product that we are going to stick with it and do what it takes.
There’s a definite kind of MicroSoft smarts. How do you define it?
BALLMER: Mike Murray [Microsoft’s personnel chief] says it’s like math camp. A bunch of people who really think there’s a correct answer, and want to get the correct answer. Everything’s like a math problem. It’s a kind of quant-jock intelligence. And the kinds of questions we ask [in a job interview] reinforces that. Can you do structured problem-solving?
What are some examples?
GATES: Well, we always switch them around because otherwise, everybody knows what they’re going to be.
BALLMER: For years, I used this one quite a bit. I’d ask people to pick a number between one and a hundred. You get it on the first guess, I give you five bucks. Takes you two guesses, I give you four. Three, two, one, zero. Then you pay me a buck, you pay me two. Do you want to play or not?
GATES: And you’re telling them if they’re high or low?
BALLMER: I tell you high, low on your guess. Do you want to play or not?
GATES: And getting the right answer isn’t the key thing, if the person can think about it in the right way.
BALLMER: You want to see that people can think in a disciplined, rational way. Although I will admit that someone once wrote down that this has an expected value of negative 21 cents as soon as I finished talking. [Looks over at Gates, who’s started to jot down numbers on a piece of paper.] Look he’s working on the problem!
GATES: Just trying to get 21 cents, that’s all.
I’ve heard Steve say the worst thing you can do with Bill is not admit problems.
GATES: Yeah. If the program is too slow or we’re losing money doing something or somebody else has a better idea. Bad news is actionable. And so bad news has to travel fast. So whenever they send me financial reports, it’s those red numbers that I double-click on. When you get people used to doing very well, there’s a certain laziness that creeps in. It’s “Send e-mail about this great win we’ve just had!”
Have you ever argued about whether you may be getting too greedy for market share? About the risk of a backlash?
BALLMER: There’s no sin in doing great products, offering them at a good price, working aggressively. The place where we have more disagreement is about how much to do about getting into, quote, “new fields” for us. I tend to be more conservative about wanting to get into new areas. Take interactive media. Which of these new investments should we do?
GATES: But we are doing interactive content. We’re doing enough of those things, it’s fair to say some will not be successful. And even success we won’t measure in the same kind of market shares that measure a successful software product.
BALLMER: Some of these things are a real pain in the neck, frankly, for me day to day. I don’t manage them, but I have all these customers who wonder before they buy the base products-the banks we have to tell “We’re not a bank, we’re not a bank! We’d like to sell you some NT servers!”
GATES: That one is just people getting confused…
BALLMER: But when you get into a business, how can you not try and make your products as good as possible and get all the customers? It’s legal, it’s proper, it’s capitalistic, it’s right, it’s good for customers. Do we actually sit and say “Let’s try not to get market share.” The answer is no, not a conversation that we’ve entered into.
Are there any other companies that could still come along and eat your lunch?
GATES: Netscape is still a scare. And perhaps the network computer. Our top priority this year is “manageability,” making your software transparent. If we don’t do that well, it would restrict growth and create an opportunity for somebody.
BALLMER: It’s still ludicrous that nobody’s ever made a run at us by making UNIX a popular server platform on PCs. It’s almost too late now.
GATES: It’s too late.
BALLMER: OK, it’s either too late or almost too late. But there’s been a hole.
GATES: If somebody does speech recognition better and sooner, you could build a device that was incompatible and yet would still be attractive. And it’s a question of when, not if. And that’s partly why we’ve got so much money into our research activities including those key areas. But you have to attack before we recognize the problem. On manageability-there isn’t a hole there now because we’ve got the pieces going…
BALLMER: [Rivals Scott] McNealy [of Sun Microsystems] and [Larry] Ellison [of Oracle] botched that up in my view. They didn’t have to get us so focused. We would have been 60 percent focused on this manageability thing anyway. Now we’re 110 percent focused.
The corporate market and NT–what do you need to do to make that successful?
GATES: Well, time is on our side because Intel is going to keep making better chips. The applications are coming along. We just need to do what is straightforward for us, listening to customers about what features are missing and continuing to enhance the service capacity. NT-5 is a big release for us. That goes to beta later this year.
BALLMER: I actually think our bigger opportunity is in small businesses who may not even have servers. We have some exciting work going on to try to target server product towards the small businesses of up to maybe 25 or 50 employees.
GATES: The power of integration is important to all customers because it simplifies things and it lets them get a rich set of services. It’s easiest to appreciate when you get down to a small business and you say OK, this guy doesn’t want to think about whose SQL is faster and whose ATTP is better and who has SMTP mail. Why not just give him a box that does all of stuff without ever exposing those acronyms?
The Comcast deal: what’s the best- and the worst-case scenario for what that investment does for the company?
GATES: Well, it’s interesting, nobody’s ever asked me before where our cash was invested. You know, were we getting a good rate of return? We think the expected rate of return is higher than the municipal bonds that we otherwise would have had. Best case? When we bought Web TV, it was an acceleration of our work on connected TV. And when you have a high-speed connection, it makes connected TV a lot better than having to wait to dial up and having a slow-speed connection. We think connected TV over the next five years will become a big thing. So it’s a nonvoting thing where we’ve agreed to partner up to prove that connected TV can really work.
BALLMER: You could look at it as Web TV on steroids. You know, you’re watching the baseball game and then you want to get the statistics on Ken Griffey. Or you’re watching something and you want to shell out and buy something that you see.
Last fall Bill started talking more publicly about Steve and his role in the company, that you considered him the No. 2 two and that If anything happened to you, Steve would take over. Is that something you talked about?
GATES: No. I think it’s a phenomenal business partnership. I wouldn’t enjoy my job like I do flit wasn’t for how much fun Steve and I have brainstorming things. And within the company, everybody has understood that we work very closely together and have a very common view of where we want to go. Externally, people tend to identify the company with one person. It’s a natural thing so I’ve had all of the–mostly the minuses, but the pluses of that.
BALLMER: There’s been a lot of minuses, in my view, which I thank you for taking.
Is it true that you critique each other’s schedules?
GATES: Yeah, I have Steve look at my calendar. It’s a conversation we have at least 10 times a year: “I’m feeling overloaded again. I wonder if I’m spending my time the right way?” And so Steve will get my calendar and flip through it and say, “Did you really need to do this speech? Did you need to meet with these guys?”
BALLMER: “Is it PR?”
GATES: Too many interviews, that’s right. Could you have done a 10-minute interview instead of a 50-minute interview?
Together you guys are worth what, more than $40 billion? That’s a staggering amount. What keeps you going?
BALLMER: Pretty much the same thing that kept us going before we were worth whatever the number is today. A) it’s fun to build stuff that people like, appreciate and changes the way they work and live. B) it’s fun. Business is in part a game. You’re seeing if you can outthink, outwit the other guy. It’s a puzzle; it’s a chess game. I mean, heck, I like playing chess. I’m bad at it, but I like playing basketball. Bill has other things he enjoys. Business, we’re actually pretty good at.
Read It All on AOL You can read a full transcript of this two-hour interview on Newsweek Interactive on A0L (keyword: Newsweek).
