Steve Ballmer: Microsoft Business Summit 2005

Remarks by Steve Ballmer, CEO, Microsoft Corporation
Microsoft Business Summit 2005
“Together, We Build Business”
Redmond, Washington
September 7, 2005

ANNOUNCER: Ladies and gentlemen, please welcome Chief Executive Officer Steve Ballmer.

STEVE BALLMER: Well, thanks. It’s an honor and a privilege to have a chance to be here with you today. I want to start by thanking so many folks for coming, and spending the time with us here. I know for those partners, those customers, and those press folks with us today, it is a huge commitment of time and energy, and particularly on a day when a lot of what we’re talking about is how time means money, I want to say thanks very much for spending the time with us today.

I want to try to provide a little bit of what I would call high-level context to some of what you’ve heard today during the course of the next 30 minutes or so, and then have a chance to address some of the questions that came in as folks had a chance to register today. We’ve tried to pick questions which we don’t think we yet answered this morning, but we have very much a mix of things to talk about. I want to go back in some senses to what I would call the origins of this generation of focus at Microsoft on small and medium-sized companies. If I go back historically, and I really date it to about 1992, I remember moving over to sales, and I brought some of the smart marketing guys with me, and we really just had “the sales force.” And the sales force sold to everybody. It didn’t matter whether it was a retailer, a partner, a large enterprise, midmarket, we were just very uniform in our thinking. And within a couple of years, some of these very bright folks figured out that there really were very big differences in the needs, the buying patterns, the partners, the way IT was done, across the spectrum of customers that we served. And we said, look, we’ve got to get smarter about this, and we put in place dedicated sales forces, if you will, focusing on the needs of our larger enterprise customers, and certain key partners who serve them, midmarket and small customers, and the partners who serve them, and then the consumer being focused in largely at that time by the retailers.

So, just about another year or two, and we figured out that it wasn’t just a question of the way we sold to people, and the way we presented ourselves, and called on people, but frankly these customers had very different needs in terms of licensing and pricing. And so, we spent several years really focusing in on that. You’ll hear us focus in on that again a little bit more. Unfortunately, we’ve learned in the licensing area enough is never good enough. We have to continue to innovate, hopefully more steps forward than steps back, although we occasionally have a little step back there, too. But we engaged in innovation that was around the pricing and licensing models and, in fact, also around the support model, because the support needs of a large enterprise with perhaps hundreds or thousands of IT people is really quite different than what we see amongst our smaller and midmarket customers. So, by about 2000 or 2001, we had, I won’t say exhausted, but we had spent a good period of time really focusing in on a number of the differences between small and midmarkets, and large enterprise customers, but it wasn’t really until the late ’90s, early 2000, that we took what I would call the perhaps most important step back and asked, what really needs to be different in our products themselves. How do the products need to change and morphed to be more appropriate and smarter for businesses of all sizes. Some things are in common. But there are things that we can do if we focus that are unique to the small, to the midmarkets, and for the large enterprise customers.

The other question we asked ourselves at the same time is, how do we want to evolve our portfolio of products? What additional product areas, product lines, businesses, should Microsoft be in in order to add value, and have the right kind of presence with these customers of various sizes?

So, you look at the large enterprise, and I’m going to start there for a second, with hundreds of people in IT, and IT budgets that can run literally to billions of dollars, a lot of specialization of role, et cetera, a lot of ability to handle specific products with complexity for people in different functions, and we said, look, we can’t be all things to all people in the large enterprise, but we are going to be the core provider of infrastructure for information workers, IT pros and developers in those large enterprises.

Complete Supplier to Midmarket

But, when we took at look at the midmarket and small business, we said, hey, look, these are different. And today I want to really focus in on some of the differences and thoughts we had about the midmarket customers. You’ve heard a little bit from Orlando and others about the size of this market globally, the number of customers that are engaged, the level of spending and effort and energy people put into IT, but what was very clear to us, certainly five-six years ago was that perhaps the midmarket customer is the least well-served customer across this spectrum of people involved in IT. The small business has its own set of issues, but, in a sense, small businesses almost by definition are less complex, want less complexity, and have fewer expectations. The midmarket customers, we feel like has to compete, and is challenged by the large enterprise, but is also challenged by complexity and scale to find solutions that are very appropriate.

So, we put our R&D minds to work on this problem. The first product we came out with, I would say, was our small business server product, which is really targeted at companies with one server. Some of them are small, and some of them are actually midmarket size, but it’s really a customer who is, by and large, going to have one server. Not zero servers, which you find in many very small businesses, and not four, five, six, 10 servers, which would not be atypical to find in a midmarket-style account. Some small businesses might have five servers, and some midmarket customers, as I said, might have one, but it was a product that tried to extend innovation, and thinking about the very specific things from a product perspective that are going on in these customers of different sizes.

We got off to a reasonable start, not quite as fast as I anticipated when we brought that product to market, but over the last five-six years, we’ve really seen it ramp up. And so we were imbued five years, even, six years ago, with this sense that said, if we applied our R&D innovation capacity in addition to sales, marketing, support, licensing, there was an opportunity to make a real difference, and that part of that difference would have to come from evolution in the product portfolio. And we said, look, probably the best thing we can aspire to, and I’m not going to try to claim we are there today, but I am going to claim we aspire to today, is to try to be a reasonably complete, I’m not going to say complete because that would imply all things to all people at all times, but a much more complete supplier to the midmarket customer. If we want to serve the midmarket customer well, we have to allow them to do amazing things with IT, the kinds of things people expect from large enterprises. We have to give them a level of support that’s appropriate, and yet we’ve got to take out a whole level of complexity through the process. Part of that would come through the way we engineered individual products, products like Office, products like our Windows Server products, but some of that was frankly going to come by putting together a broader product set, and have that broader product set really be engineered to take overall complexity out for the midmarket customer, and their IT people.

A typical midmarket customer, we did this about two-and-a-half, three years ago, we sent out 80 people, we said, just go live with midmarket customers for a month, and then report back what you saw. That’s typically a company that might have anyplace between three and ten IT people, might have anyplace between a couple and ten or so servers, although you’ll find some people whose business is IT in the midmarkets who would have many more servers, take people who run Web sites as a good example. If you stop to think about it, what does it mean to have five or six IT people, it basically means you have one person who takes care of the desk, you have one person who takes care of the core application, the people in the room, I see some people smiling around here, you’ve got one person who is the boss, and the boss here doesn’t kind of mean traditional CIO, it means person who will do anything that needs to be done in case the rest of the department can’t get it done. You probably have one person who takes care of the rest of the servers, and that would be not an atypical configuration. These midmarket companies were complicated, they weren’t centralized, they had branch offices, they had remote connectivity needs, and we said to ourselves, we’ve got to engineer a product line that makes sense.

We said, look, if we’re really going to be a more complete supplier to this group of customers, part of what we need to do is be more deeply involved in the business of the business, which means we really do need to get into the ERP and CRM realms. And so we made the decision and the commitment for the midmarkets, and below customer base for the small customers as well, that we were going to go after ERP and CRM. We were going to do it with energy, enthusiasm, innovation, and probably most importantly a hallmark of Microsoft, a certain patience and tenacity, so that if we didn’t get it right at first, we could keep working it, and working it, and working it, because our customers have come to depend on us to get things right. If not initially, keep after it. And a lot of the long-term investments customers have made with us, I think they have been very proud of. Some of them in six months, some of them took a couple, three years before they got proud of them. We wanted to make sure that the customers in this space particularly, where IT budgets are lean, where the staff is lean, that people knew that they could count on us for the long-term. So, this is a long-term important area of investment to us.

We said, look, we’re going to take a look at what it takes to evolve Windows and Office, and our server product line. We’re going to expand, first by the acquisition of Great Plains, then we said we need more of a footprint in connection with our customers in Europe, and we wanted to get there quickly, Europe and elsewhere in the world, so we made the acquisition of Navision. And we said, now, we’ve got to get to work. We’ve got to integrate these companies, and we’ve got to figure out how to make this all work. We’ve got to get the best of all words in terms of getting things kind of more unified, simpler, take out cost and complexity, and make the whole bigger than the sum of the parts.

So, when we show these snazzy little videos, like we did before I came on stage, that talk about empowering people, it’s because we see our vision for the midmarket customer as essentially that: how do we take all of these parts of technology and make them work harder for the IT person, and for the various other roles, the various information workers, the various task workers in these businesses, and bring some of the simplicity of a product like Microsoft Office or Microsoft Windows to the broad IT needs of the midmarket customer, and do that in a holistic way, in a sense, the way we have with Office for the information worker, and the way we have with the Windows Server suite of products for the IT person in the enterprise data center.

So we stepped up, as I said, to continual investment. We had learning we had to do about big acquisitions. We’d never done a big – I know some of you may still think we haven’t done a big acquisition. But, today, by our standards the biggest things we ever bought were Great Plains and Navision. We had a little bit of a learning curve on how to get them integrated. It turns out it’s a little more complicated than I guess I might have anticipated. Here we sit in the year 2005, I can tell you we’re completely through that learning experience at this stage, but it lasted much longer than it probably ought to have.

And last, we had to continue to build our understanding of the customer segment. I told you how it kind of started in ’92 or so, but we still have a lot to learn about how to apply the innovation. What does it mean to try to bring together the best of products like Solomon and Axapta, and Navision and Great Plains. We had to learn about that. How do we bring the power of a product like Office to bear against a line of business processes, what did it mean to involve office? What did it mean to evolve these ERP and CRM products to have the simplicity of Office? So there was a lot of additional customer understanding.

Turning the Corner

Today I think we are through those acquisitions, you see that some in the product naming, and some of the product directions you had a chance to hear. I think certainly we do think we do have a deep customer understanding. Today we can talk to you about what it means now to do a midmarket server, as we described to you today. We can talk about what it means to involve Office in line-of-business processes in a way that simplifies life for the midmarket IT person, and for the end user who wants that access to that line-of-business data.

In a large enterprise they’ll take the line-of-business data and throw it off into a data warehouse, and there’s IT people taking care of all that. In the midmarket business it needs to just kind of work right out of the box, the connection between the user and the enterprise data that’s important. We’re engineering that into Office, into our new Dynamics product line, et cetera.

So I think the kinds of things we’re announcing today really fit into that context of breakthroughs, and at the same time I’ll tell you this is in a sense the kickoff of the next generation of innovation, where you can really see our full R&D innovative value applied to the specific interests of the midmarket customer, through ERP, through CRM, through Office, as well as through the server infrastructure, and the services themselves. I’ll talk a little bit more about this FrontBridge acquisition that we’ve made, but the midmarket customer will be perhaps on the leading edge, in my opinion, of business customers in adopting Internet-based online services to take  to even extend further, taking out cost and complexity for midmarket IT. And the company, FrontBridge, that we bought, that’s where it has its primary traction is in really serving midmarket-type of customers.

Today you’ve heard about innovation in a number of different areas. In the next version of Office, Office “12,” we’ve shown you some things, we’ve talked about some things, but what we’ve really done is work on the workflow and business intelligence capabilities in Office, in a way that Office can be really made part of these line-of-business processes. And our Dynamics product line in Office, we will make those things just work out of the box. We will make them work with effort and energy, in some cases with other ERP and CRM systems, blah, blah, blah, but our stuff you can count on to just work out of the box.

We talked to you about this midmarket server that we codenamed “Centro,” that’s a product that benefits from all of our learning about the midmarket, and all of our experience with Small Business Server. I will tell you that when we first started talking about this product, I had kind of one conceptualization in my head of what it was going to be. And after our people fanned out and talked about various product concepts with various folks in this room and many others, we wound up with a product that doesn’t look anything like that which I conceptualized. But, the feedback that we’ve gotten is we’ve concept tested, and we’ll start the beta test process with midmarket customers, has been phenomenal, in terms of what it can mean for extending functionality, taking out cost, and taking out complexity.

I started with the following basic proposition. We have too many products that we’re trying to sell to too few people in the midmarket. That was the basic proposition. If you only have five people on an IT staff, or six people, how can they possibly understand 10 server products and 20 client products? So I had sort of one view of where that took us. I’ve got to say, I started thinking that way when I said, why can’t we sell more of our products to the midmarket, so it was borne of a view of growth and revenue opportunities. And I think what we’ve come up with is something that really is a win-win. It’s a way for us to get more of our valuable technology in your hands, and a way for that technology to deliver more value in your businesses.

Then last, but certainly not least, is the work that we’ve done in Dynamics, to think about how we put enterprise systems to work for people. Part of that is the integration with Office and the user experience. Part of that is this very role-based modeling that we’ve done to define those products, and the roles that various people who work for you play in those product sets.

Today we’re also introducing our new Open Value license. If you look at who has the greatest set of issues, concerns and challenges with our licensing terms and conditions, it’s the midmarket customer. And there are two essential elements, from my perspective, to what we’re trying to do with Open Value. No. 1, we are trying to be aggressive in permitting people to use our software, essentially at lower cost, that’s number one. And No. 2, I think as importantly, we’re trying to simplify the process of license acquisition, and maintaining kind of a legal position with respect to software licensing, including things like simpler contracts, easier to understand what you need, better tools to help you analyze what licenses you need, easier tools to help you understand what licenses you’ve already bought. We’ll keep a database, as long as you’re buying our Open Value licenses online that helps you track and understand your license position, whether you’re legal, et cetera.

Last, but not least, is the service aspect that I mentioned. I talked a little bit about FrontBridge, I could talk about Live Meeting. But, I think increasingly from a midmarket customer’s perspective, the best piece of software is the piece of software that they can subscribe to, as opposed to implement. Not everything will fall into that category. You may want to do a customization of our Dynamics products, ERP or CRM, that you want to have your own instance. You may want to host, because you want to be able to control not only customization, but certain operational characteristics of our products. But, I do think midmarket customers are the most open to subscribing to and using hosted business services up in the cloud, and with FrontBridge, with Live Meeting, and with a set of other things that you’ll see us bring to market over the course of the next year or two years, you’ll see us really push to allow our customers, if you will, to have it their way in the midmarket, on-premise or off-premise implementations of a variety of different solutions.

The other key thing I want to come back to is something that I teed up, that we started on much earlier back in the ’90s, and that’s this notion of what kind of a relationship do you have with a midmarket customer. With a small business, in a sense, it’s simpler. There are so many, many millions of small businesses, you can’t confuse yourself that you’re going to have a very personalized relationship, you’re lucky if you know their name and what they’ve bought. Large enterprises demand a very personalized relationship. The midmarket customer, again, is someplace in between.

So we’ve been seeking to have relatively personalized relationships, and still recognizing the fact that there are millions of midmarket businesses. Many of those midmarket businesses would spend nothing to several thousands of dollars a year with us. The typical midmarket IT budget could be anyplace from $15,000 or $20,000  well, without an IT person $15,000 or $20,000 a year, but typically $50,000 on up to maybe half a million or a million bucks. You’ll get some very large midmarket customers that would spend even more than that, total IT, not with Microsoft, but the total IT budget.

So we said there are two things we’ve got to do to re-engineer our relationship. No. 1, we do need a broader product portfolio. If we’re going to have an extensive relationship with a midmarket customer, it is easier for them and for us if that’s across a broader product set. But, No. 2, we are going to invest in the partnerships, and we are going to invest in the tools to allow us to have a personalized connection with over a million different customers.

So we’ve done work on our own internal IT systems, our own internal Web sites, whether it’s things we’re doing in license reporting, or Web-based seminars, or our own CRM implementation, or the work that we’ve done to create a database of our partners and put it online, so that our midmarket customers can find the people who’ve got the solutions specific to their needs. We put a lot of energy into really thinking through how to have personalized, human, but a broad scale relationship with this very important customer base.

Deep Midmarket Commitment

The cornerstones I’ve talked about of this commitment. One thing I think is important for me to emphasize, and I’m never sure whether it’s the right thing to do or the wrong thing to do, but I see great opportunity for revenue growth for Microsoft coming from midmarket customers. If you’re a midmarket customer you might sit there and say, gee, I’m not sure I really want to hear that, do I? At the end of the day I think you really do. It’s not because we think there’s some great opportunity to continually increase the prices to the midmarket customers, but there’s an opportunity to better serve, to deliver more value, to touch more midmarket customers, to provide a broader product set, more features, more capabilities, to broaden our portfolio, whether it’s security, or ERP, or CRM, in a way that lets us serve that midmarket set of customers better.

The fact of the matter is, if you take a look at all of the data I see, customer feedback, I think there is, perhaps in a revenue sense, as much opportunity for our company in the midmarket as any place in the world. I think if you’re a midmarket customer you put your hat on and say, that’s good. My economic interests of good service matches their economic interests. They, Microsoft, will do their very best work, their best R&D, their best sales, their best customer support, their best licensing, the best job they know how to do to take complexity out, because I am a valued customer, I am part of what Microsoft has to do to make not only innovation, but its business, and financial, and other goals.

So we’re excited about this midmarket. I personally would say I’ve probably spent more time trying to understand the midmarket than I have any other customer segment over my years at Microsoft. And it’s not complete magic, on the other hand, it requires a little more work than it does to understand either the consumer or the enterprise. Consumer, everybody will tell you what you need to do, just ask them if you want, the enterprise, they’ll tell you what to do, whether you can do it or want to do it or not, no enterprise customer is going to let you off the hook. And we have enterprise customers parading through here literally daily to talk to us, and we’re out there talking to them, and we’re getting the feedback.

The midmarket customer you’ve got to want to connect with. As I look and compare to other participants in our industry, I don’t think it’s just me, I think our company has worked harder to understand the needs, the product needs, the product portfolio needs, the innovation needs, the complexity needs, the pricing needs than any other business in the market bar none. We actually have the biggest business in our industry selling to midmarket companies, and yet we continue to see incredible opportunities to innovate, to add value, and to grow.

With that, I’m going to wrap. It’s been a pleasure to make some of these formal comments to you, and I know Orlando Ayala is looking forward to joining me on stage with some of the questions that you submitted at registration. So I’ll end by saying, thanks.

If you are either a press person or a customer, or a partner, with something on your mind, I am [email protected]. As some of the customers in the room know, I’m not always 24-hour turnaround on e-mail, but at the same time I do get back on all of it. I’ll be seeing afterward a customer who I first met on e-mail when he had a negative experience with our SMS product a few years ago, and has really been a leading-edge customer in helping us push through and work through that technology. And I’ve actually had a chance to meet a number of you, first electronically, and if there are things on your mind, please feel free to ping me. Thanks, I appreciate it.


Executive Discussion

ORLANDO AYALA: Well, in this home stretch, we’re going to have some Q&A time. And, first of all, thank you very much for all that participated. It was quite active, and we really appreciate it. We promise we’re going to have you out of here by 12:15. So, we’re not going to be able to answer all the questions. I want to remind you that in the CIO Summit tomorrow, for the most part, Steve is going to do Q&A, and you’re going to share some thoughts with him. So, you will have the opportunity to ask the questions you submitted.

So, we have picked five, which I think reflect a little bit of the areas where we want to expand perhaps. Here’s the first one: If you were a CEO of a mid-sized company, what criteria and decision-making process would you use to decide the level of IT investment? What criteria would you use for choosing applications on all platforms?

STEVE BALLMER: I’ve been asked this question before, I have to say, and I always bounce back and forth between two spectrums. I want to first start and describe conceptually what the value of IT is, and how you use it, and if I was in front of the CEO of a midmarket company, I would explain those things, and walk them through the various business processes they have, and explain to them where IT can add value. And I think I’ve actually had lots of opportunity to do that with midmarket customers around the globe, with friends of mine who run midmarket companies here in Seattle, and sometimes that’s a very satisfying thing for me. Sometimes it’s actually even very satisfying for the customer I’m talking to. And sometimes people’s real question is, how much money should I be spending? Is my IT guy asking for too much or too little?

And, you know, I saw some bouncing heads there, and I don’t know where the questioner comes from precisely, but I think both of those are actually important things to be able to do with a midmarket CEO. I think most CEOs, not just midmarket, you see the same thing in large enterprises, really don’t have a deep appreciation for the value of information technology, and perhaps more importantly what information technology can mean in transforming a business.

In larger businesses, because of the complexity of running them, there’s a basic assumption that, oh, gosh, we’re going to spend anyway. In a midmarket company, where the IT budget might only be a few hundred thousand dollars, there’s a very different kind of dialogue. So, I do tend to try to walk through basically processes, not just ERP and CRM, but really business intelligence. Is the CEO equipping various people who need information with the information they need to serve their customers, to deliver their products, to provide the customer support and service, and show a few examples which our team is happy to help you show of things that people can do that are surprising to CEOs.

On the other hand, I actually do tend to get some numbers. And people want to say, OK, here it is, there’s a magic percentage of revenue that you’re supposed to spend on IT. And then people stop and say, well, I know it’s not a magic percentage of revenue, because distribution companies, retail companies, they spend less as a percentage of revenue than accounting companies. Some people say, ah-ha, it’s going to be a percentage of revenue that varies by industry. And then you get into all the sort of complicated stuff.

I say, no, it’s simpler than that. It’s just like our product stuff, it’s role-based, just think about it this way, every worker in your company, put them into two buckets. I would only put them into one bucket myself, because I love the power of information, but for purposes of the discussion with your CEO, say we have people who we want to endow deeply with information tools, and people who we are happy to endow less well with information tools. It’s a tough thing for me to say, but, OK, it happens. Some people like to use the words “information workers” and “task workers.” I almost resent the terminology because it implies everybody is not an information [worker], but the guy on the loading dock who is deciding whether to accept some goods, file a claim, do whatever, is as much an information worker at that moment in time as anybody else in the business. The customer service worker in a branch who is trying to help that customer needs to know something is as much an information worker. Let’s just say, high end and low end. You know, the low end worker, if you’re not spending $500 a year or $1,000 a year to support those workers, I would ask a question. For the high-end worker, if you’re all in, you’re not spending several thousand dollars a year to support that worker, I’d ask the question. At Microsoft, we are spending an order of magnitude 15-16,000 dollars per year, per person, to support every worker at Microsoft. And that doesn’t include If you include, the number goes up, but, of course, is primarily driven by the number of touches and hits externally, and it’s one of the top few Web sites in the world.

So, instead of doing it by business, or by magic percent, I would encourage you to say, look, if we’re not spending $500 a head on our lower end heads and at least $2,000 a year on our higher end heads, we’re probably low. And if you’re spending $5,000 a year on everybody who works for you, you’re definitely at the leading edge, shall we say. Again, all this assumes that a big customer facing Web site is not your primary business.


ORLANDO AYALA: The second question was actually asked multiple times by people and that reflects the interest of people. There’s a lot of news in the marketplace about Linux. How should we think about Linux in our long-term IT strategy?

STEVE BALLMER: It’s me, “You don’t have to” would be my answer. More seriously, you know, I think two things are true. No. 1, it is great for every company, including our own, to have competition, and we certainly feel competition with Linux. That’s our primary relationship with Linux is as a competitor. And I know that the competition with Linux motivates us and drives us to think creatively. It encourages us to innovate around things like Small Business Server, and the Midmarket Server. It causes us to think through how do we provide a more complete and interesting value proposition that includes things like security and management, et cetera, for our midmarket customers. It causes us to want to make sure we’re watching our prices, and keeping them sort of a good value, if you will, for the money.

So, I think it is fantastic. That doesn’t mean everybody – I think I have to agree it’s good for everybody to have competition, and I think if I asked you, do you really enjoy day-to-day having competitors, you’d probably say, no, I’d prefer that they not exist. But, let’s face it, it’s a good thing. And so, we compete with Linux.

How should you as IT people think about it, I think, look, every day you’ll come to work and you should make one decision and one decision only, am I getting the product and services that help me run my IT shop the best efficacy, that means output and input, for my organization? And I think if that’s the test, Microsoft products and Microsoft offers will, the lion’s, lion’s-share of the time, be the right solutions for you. If you only have five, six, seven, eight people in your IT department, it’s not clear to me how much time you want to spend trialing new things, experimenting, but sometimes maybe you should if that’s the right thing to do. It’s our job to make sure you shouldn’t have to think about Linux if you don’t want to.

Anything Linux does, we have an engineer or a set of engineers come to work and say, anything they do, we’d better make sure we do better. Right now, Linux is ahead of us, for example, in the market for high-performance computing clusters. Not a huge midmarket area of interest, but nonetheless an important one for us. And we have guys come to work every day and say, how are we going to get better at that. We have guys who come to work every day and say, how are we going to make sure that – and we do have midmarket customers who are in the hosting area, how do we make sure that for low-end Web hosting, Windows supersedes Linux in every way? We have some customers in the midmarket that say, look, it’s free, therefore it offers better value. A: It isn’t free. And B: At whatever price it winds up costing you, I’ll tell you that the lion’s-share of the time, we can show you that you’ll get lower total cost of ownership, more capabilities, and more output from a Windows-based solution.

So, we’ll give you the sales pitch in answer to the question. Let me just read it again, there’s a lot of news in the marketplace about Linux, how should we think about Linux in our long-term IT plans? I hope you’ll mostly think about it as that nice competitor that drives Microsoft to do the work that makes Microsoft our most important vendor.

ORLANDO AYALA: Here’s a question. You and I know that in the midmarket the customer segment is actually very sensitive about industry solutions, I mean, the decision-making process, the relevance of the industry they are in, it’s quite important in our research. So here’s a question, I have heard news recently about Microsoft going vertical, does this mean that you will be developing vertical flavors of your applications?

STEVE BALLMER: I would frame it a little differently. Every year we will add more capability to our Dynamics product line. Every year we’ll add more capability, it could be manufacturing capability, distribution, professional service, billing, we’ll continue to add more capability. Our basic strategy would be to add more what I would call horizontal capability, and at the same time to add more extensibility so that third parties can take our Dynamics products and extend them in ways that are really 100-percent tailored to various industries.

The market in which we have our best traction worldwide is Denmark. That may be because of the merger of Vanguard and Navision, which led to the Navision company that we acquired. Denmark’s a country of about 5 million, 5 and a half million people. We have a catalogue of over 300 vertical solutions in the Danish market coming from our partners. Some of them come from partners that write them, as applications that they sell, and some of them are just partners who have proprietary software that they’ve written that they will install along with Dynamics for various customers, 300 solutions for 5 and a half million people.

We are working to make sure that we have enough horizontal capability in Dynamics, and enough extensibility that we can have the equivalent of that catalogue available in every part of the world. I don’t even say every country, because as I tell our team here, we may need to have 20 solutions for everyone in Denmark, one for Southeast Florida, one for the Northwest, one for the New York area. We need partners who specialize even by part of the country here in the United States. So for real verticalization with depth we want to make sure we have an enabling infrastructure, and we work super well to have the broadest catalogue of very deep vertical solutions available.

I have a friend here in town who is a clothing manufacturer, and when we first bought Great Plains he said, I’m your friend, I’m going to buy Great Plains. So he trundled out, he had his people look at it, and they thought it was great, and they said, we’re not buying it. I said, why not? He says, you don’t really do a good job yet with colors and sizes. And it turns out colors and sizes turn out to be semi-complicated in the ERP world. So we really have to work. We’ve got to build on our horizontal functionality, and then we’ve got to have partners who do a wonderful job of managing the specifics of the garment manufacturing business.

Just as an example, we’ve got partners in Denmark who specialize in frozen-food distribution. It turns our frozen-food distribution, not surprisingly, is a little different than non-frozen-food distribution. And you don’t really want to screw up the difference either, you’ll be out of the frozen-food distribution business pretty quickly. So we recognize the need for verticalization, we’re pushing our products to be better, deeper in a sense across all verticals, and we’re working with the partners on a big catalogue of solutions.

ORLANDO AYALA: I know you have been pushing very hard, a true believer of this. And I just wanted to make the point that our goal is to have 12,000 of these micro-verticals in the next year available to customers locally, which is very important to the local relevance, it’s super critical. So watch us, we’re going to be working very hard on that.

Here’s the fourth question, and an interesting one. It’s kind of unique. We’re a manufacturing company with 700 employees. We’re in the process of choosing a new ERP system. SAP is the best known vendor. Microsoft has several different products in this space, why should we choose Microsoft over SAP? If we choose Microsoft, which product should we choose, what Dynamics product should we choose?

STEVE BALLMER: Well, I’d say a couple of things. The first thing I’d say, given that the person identifies that they’re in the manufacturing business, the first thing I would do is just to check quickly whether there’s a vertical solution for their specific sub-segment in our catalogue, that would take me to one of the specific Dynamics products. And if there wasn’t, I would naturally go to Dynamics AS, the product formerly known Axapta, because it is the strongest, and richest in terms of its manufacturing capabilities.

Why look at our Dynamics products versus the SAP products? In both cases SAP has got R3, they’ve got All In One, they’ve got Business One. So there are three different products from SAP to take a look at, there is our Dynamics product line to take a look at. I think when you take a look at the midmarket customer, there are two things that distinguish our product line, number one, simplicity of implementation. The time between acquisition and implementation, people talk about the SAP horror stories, months, years, multiple years, people get implementations done on our Dynamics products in weeks/months, not months/years, even in very complicated cases. That’s the kind of implementation time.

So from an IT perspective I think there’s simplicity of implementation. From an end-user perspective there’s also a simplicity in the software itself. If your folks just spend a little bit of time taking a look at our screens, our user interface, our connection with Office, the ability we give you to present business information to decision makers in your company, I think it fairly quickly looks pretty compelling versus what SAP has in the market or will have in the market any time soon. That’s despite the fact that in many ways we have a great relationship with SAP, we partner with them up in the large enterprise, but when it comes to the midmarket and connection with Office, and what we’re doing, we think in general we have a superior solution.

Each of you, for your own evaluations will go through and take a look at functionality, and function points, and I think what you’ll wind up finding is that us and our partners versus SAP and its partners, we have a broader ecosystem, and we tend to have more complete solutions than SAP has, unless you’re in a specific area where SAP is built out of a vertical solution, in which case they may have more functionality, but they’re also likely to have infinitely more complexity that came with that functionality.

ORLANDO AYALA: Just one thing, in very tight IT environment, this customer segment is always  to extend your remarks, a very important consideration is portability.

STEVE BALLMER: That’s right, portability and time to implementation, time is money certainly for our midmarket customers.

ORLANDO AYALA: Okay. Last question, has a lot of momentum. When do you think it makes sense to go hosted versus on premise? What are Microsoft’s plans about managed services?

STEVE BALLMER: As I kind of spoke to directly, and alluded to more generally in my remarks, we do believe that there will be a set of services that our midmarket customers, and small for that matter, consumers, large enterprises, right now I’m talking midmarket, customers will be interested in using or running on a hosted basis on the Internet. I talked about Exchange Services, anti-virus, anti-spam that we’ll provide through FrontBridge, and certainly there are customers who want hosted CRM solutions. We will provide details to come when appropriate. We recognize that need and you’ll see us respond to and address that need in the marketplace, and a variety of others. You’ll see us announce a variety of services for smaller and midmarket size companies over the course of the next 12 months.

When do you go hosted versus when do you go local? Today, if you’re going to do a lot of customization of any software, or much customization at all, frankly, or much integration with other systems in your business, you will almost certainly use an on-premise solution. The level of customization that is permitted is still much lower for hosted solutions, whether it’s Salesforce’s or the stuff we’ve done and are working on. There’s just a lot less customization and integration that’s possible with other systems that you run than there is in the on premise case, which is much stronger on that. If you’ve got a simple, light-weight application, if it doesn’t matter that you guarantee yourself the kind of reliability and availability that you want to take personal accountability for, I think a hosted solution can look good for a variety of different things.

So I certainly see a balance. I think right now it’s more the leading edge customers that are looking at some of these hosted solutions, and there’s a reason why it’s the leading edge customers, because they still have things that they need to prove, but we’re certainly on our front working very hard in that direction, and we expect to give a very effective run for its money, by having both on premise and hosted solutions over time.

ORLANDO AYALA: All right. This concludes our Q&A, and again, I’ll remind you tomorrow you’re going to have the opportunity to engage with Steve on the live questions. So I want to thank you, Steve, for the time. Thanks.

STEVE BALLMER: Thank you very much.