Steve Ballmer Speech Transcript – The Microsoft Analyst Summit 97

Remarks by Steve Ballmer, Executive Vice President, Sales and Support, Worldwide Business Strategy Group

Thursday, July 24, 1998, Seattle, WA

MR. Ballmer: Before I get started, I want to address a couple of the issues that people raised with me over the break and the lunch.The first is an apology, I guess.I think most people thought, with the earnings release out last week, that it would be kind of a calm and cool and relaxed morning before this analysts’ meeting.And I gather that my comments, which were reported on today in The Wall Street Journal, caused people a level of consternation this morning, for which I certainly am apologetic.

The other thing is just a comment, and I thought I’d address it square on.People noted that I’d remarked quite vigorously about our stock being overvalued.And people asked me, “If it’s overvalued, why don’t you sell, Steve?”And there was this sort of deep-seated tinge of “You hypocrite, you lousy hypocrite.”

(Laughter.)

And the truth of the matter is, despite how negative I’m going to be for the next 44 minutes and 6 seconds, I love this company.

(Laughter.)

And nobody should get confused about that.I love this company, but we’ll have a chance to share with you some of the issues before us.

The first thing I want to do is comment a little bit about the last 12 months, and particularly some of the things that we’re seeing going on in the corporate state.Bill hit on this word, I don’t know, probably three, four months ago.He started talking about things inside the “digital nervous system,” and how PCs can grow to form the foundation for a digital nervous system inside companies that let the companies see and communicate and react, empowering themselves in a wide variety of ways.

It’s a great expression and one that you ought to expect to hear a lot more from us because it really resonates in terms of what we see customers doing and customers wanting to do with our products.The PC has really come full circle in its evolution from being simply a platform for personal productivity to really being a backbone building block that lets these customers not only get competitive information, plan, react to unplanned events, and move forward in very strategic and smart ways.

It’s not like the mini-computers, which were simply a device that knew how to move material and process paper.It’s really a device that gets involved in the analysis, the planning, the thinking, the tracking, the delivery of knowledge, and the training of employees inside organizations.And we’ve seen that happen fairly broadly around the world.

A good digital nervous system, of course, depends very heavily on having PCs quite broadly in the hands of the employees inside the company, but it also depends on good productivity applications that let people do analysis; that let people communicate not only inside the company, but with suppliers and customers outside the company.It depends very heavily on a great network, and a great messaging and communications system to allow people to move information around and structure it in an ad hoc way.

It depends on Internet connectivity in order to be able to connect with, see and touch customers.We certainly have seen that in our own site, Microsoft.com.The value of that as a tool not only for commerce, but really for staying in touch with the pulse and thinking of our customers is incredible.It depends upon line of business applications which are very rich not only at running the company, not in doing the business of taking the orders and issuing purchase orders, etc., but also line of business applications that integrate well together and that seed very powerful analytic systems that let people really know what’s going on in the business.

One of the things that’s fascinating as you talk to CEOs, is that CEOs in most companies are interested in line of business applications because the business doesn’t proceed without them.But they’re vitally interested in the analytic systems, the ones that really let them go in and ask questions about what their customers are interested in, what’s really selling, what’s really happening.And we’ve been very impressed about the kinds of things customers have done building from Windows and Office and NT Server and Back Office to really start the process of building world-class digital nervous systems.

We’re working on that inside our own company in the way Bob described.We’re working on making it easier and easier, together with partners, for our customers to see how to do that with our products.Here’s just some of the industries and some of the customers with whom we’re doing this kind of leading-edge work: AIG and City Bank in the finance industry; Dayton-Hudson and the way they’re re-automating their retail operations and doing decision support off of that is a fantastic case; Columbia HCA, same kinds of examples in terms of communications, customer connection, decision support; the Social Security Administration; the UK army; Osaka Gas, the largest gas utility company and gas utility in the world; British Telecom; the Mirage Hotels; the Fairfax County Public Schools in Fairfax, Virginia; Indiana University; Boeing; Johnson & Johnson.

So across a wide variety of industries, there are customers who are reaching out and embracing the PC as a building block for these digital nervous systems that help them analyze, plan, think, operate, communicate, train, and educate.And it’s just a fantastic thing to see happening.But it really has happened in this last 12 months, where we’ve had the products out for a while, there’s a trained skill base of people who understand products like NT Server and Exchange, SQL Server and Office, and the momentum has really begun to build.

What we need to do now, of course, is to harness that momentum, to really make it easier and easier for customers to think through how they will build digital nervous systems inside their own organizations.

A digital nervous system is a very broad thing to think about in your company.How can it help the development of a product?How can it help your salespeople, your operations people, marketing people?How can it help the G & A people, the customer service people?There’s a lot of different things people do, analysis, tracking.And of course, you have to have a foundation of products that makes life reasonable and manageable for the people running the networks, the data centers, the help desks, the PC operations, et cetera.

We’re in a fortunate position.Because we grew up with the PC, we can help companies put in digital nervous systems, not only in the very largest organizations in the world, but also helping small and even tiny businesses have these digital nervous systems.We’re bringing to market this year a product we call the Small Business Server, which is squarely targeted at being a foundation building-block that our channel partners can take, and instead of putting all of their time into configuration and installation, we’re trying to free up time so they can help their small business customers implement digital nervous systems inside their own organizations.

We’re phenomenally excited about the opportunities.We’re talking a lot about where revenue may come from over the next year.We’re talking with the home market.Pete talked a little bit about that earlier.But certainly the big growth opportunity for us in the next 12 months is in business.And this notion of helping companies build digital nervous systems and keeping them current and competitive is an incredible opportunity for us across the breadth of company size.

Now, there are a lot of competitors who in one way, shape or form will try to deny us that opportunity.I chose to list here just six of them that I think about every day.Not a day goes by where we’re not thinking,
“what about Netscape, what about Sun, what about Oracle, IBM, Novell, Corel?What’s going on?”

And I had a funny question last night at the reception after our Windows day.Some folks asked me, “You didn’t mention Netscape much today.You must have forgotten about Netscape.Are they no longer an important competitor?”

I was horrified.How can somebody leave here thinking that Netscape is still not very much burned into our brains — (laughter) — as a competitive force, because they in their own way are trying to tell a story about some components that companies might use in these digital nervous systems.

The thing we can do uniquely, none of our competitors come close to having a broad enough product line to actually help a company put in a digital nervous system.Oracle’s got a database.They have no tools, really, for communications or analysis.IBM’s got groupware and messaging products, not much for personal productivity and decision analysis anymore, and they certainly don’t have very much at this stage in terms of helping with the core operational backbone.Novell, Corel, Netscape, they all target very, very thin slices, as does Sun.

So we have a unique opportunity through the breadth and integration of our product line, competitively in business, against all of these guys.And so they’ll attack us with their point solutions, and we will defend on a point-by-point basis.But the breadth and integration of our product line should serve as our strongest asset in terms of really working partnership with the channel and with the customers to implement these digital nervous systems inside their organizations.

We have some very good competitors.Some of our competitors have had bad years, but most of them had pretty good years.Netscape, Sun, Oracle, IBM certainly have all had impressive years in one way or another, but all of them have their own — what shall I say? — have their own Achilles’ heels as well.And there’s no competition that we find ourselves in today where we don’t feel ourselves not only very challenged, but also very well poised to make progress in the next 12 months.We feel poised to make progress even more against Netscape.

Somebody asked me at lunch, “What about Oracle?Are you waiting to take them on till SQL 7?”No way.What about Sun?What about Notes?We have the product line today in each and every case to turn on the competitive pressure vis-a-vis these sets of people.

Now, there’s going to be a lot of challenges for us both in terms of serving these customers well and in competing successfully with these competitors.I want to start in the enterprise and talk about our challenges there they’re more discussed.We have very good entrenched competition in the enterprise.IBM and Oracle, even to some degree Sun, are really thought of as enterprise players, both IBM with Notes and Oracle with its database have fairly entrenched positions.

Everybody now is very focused in on total cost of ownership.I think because of the digital nervous system and its appeal to companies, we can talk to them not only about the cost issues in the way that Paul Maritz described, but also the value of the PC and the value of that PC infrastructure.

We do have a liability today that we cover fewer enterprise accounts then either IBM or Oracle.It’s hard to get precise numbers, but I guess both of those companies really cover something between 8,000 and 10,000 accounts worldwide.We’ve been covering about 2,500 accounts, and I’m going to talk about some investment to really broaden out our present cross-enterprise accounts, and I’m going to talk about some investments in deepening our support relationship with all of our enterprise accounts.

Windows growth.

We’ll talk a little bit about the revenue impact, but everybody should understand, there’s still a very large remaining 16-bit Windows-installed base, which poses a certain set of challenges as we go out and talk to application developers about leveraging our 32-bit platform.And we do have a number of accounts where we see NCs being piloted.This is a critical year for us.We have to demonstrate the cost and value benefits of the personal computer versus the NC in these pilots.Otherwise the NC does have the opportunity to come in and cannibalize the PC base.

Some people say, no, no, no, no, Steve, the NC is only about cannibalizing the dumb terminal base.Well, the dumb terminal market — I’ve been doing some study of this recently — has really shrunk, and I think if the NCs catch fire at all, it is quite likely that it will at least retard PC growth rates, which is a scary proposition for us.

Maintaining Office momentum.

We’ve sold a lot of Office 97.People have had a chance to really read that in our press releases and financials.But not all of those Office 97 licenses have been deployed, and we need to really dig in and help customers get those things deployed so that they see the value of Office 97 in their digital nervous systems.That will help feed more customer adoption and more open-mindedness to entering into relationships with us in which customers do essentially maintenance type relationships.

The Back Office business.

There are challenges.Unix and the mainframe are clearly recognized as supportable, maintainable, high availability systems inside the data center.We took a big step forward on our so-called scalability day in really proving NC’s scalability.But we’re still having those lingering doubts about the other aspects of Windows NT server’s readiness for the data center, despite the fact that there are a lot of NT servers in data centers today.

There’s still a lot of inertia in the marketplace behind NetWare.My read on Novell’s numbers and Novell’s read on Novell’s numbers are quite different, but I think most of that inertia isn’t in the largest accounts, it’s really still in small and medium business, and we have a lot of work to do.

Oracle has incredible market share, and they execute very well.They execute very well from an R & D perspective and a sales perspective.I think we have some good things to offer in terms of our integration, in terms of where we’re taking our product line, in terms of the kinds of investment we’re making with our partners.But they’re a very formidable competitor.

Lotus Notes still has great mindshare, and now there are literally thousands of custom application development firms that know how to build Notes applications.When we go up against Notes head to head for an enterprise-wide messaging bid, we are much more likely to win than lose.But there are pockets in many companies of Notes applications that have built up, and many times that winds up serving the backbone and still providing the energy and inertia for the enterprise-wide standardization on Notes.There’s a lot of development expertise out there on Notes and Oracle.

And when you really look at it, while we’ve done tremendously well from a market share perspective with Windows NT server — of course, there’s a down side; it’s harder to do a lot better — but we’ve done tremendously well with Windows NT server.If you look at the business of selling server applications — the mail, the messaging, the groupware, the database, my guess is we’re probably only 10 percent or less of the server applications market.

The good news is I guess you could say there’s opportunity.The bad news is, in some sense from a revenue perspective and from a position perspective, we are still way, way behind in server applications, even though we’ve made strides in NT server itself.

The last challenge for us is to continue to expand the capacity and capability of our partners.Everybody in this room knows that whether it’s for enterprise accounts or small businesses, we leverage very heavily third-party software development companies, custom application developers, systems integrators, VARs, SP, and a broad community of people that we depend on in a stronger way than I think any of our customers do today.

And today we’re having trouble getting more and more of those people trained up.We’re putting in a lot of energy, but we need to put even more energy into going after the trained partners of our customers, of Oracle, of Sun, etc, etc.And there’s a lot of work to do to capitalize and build that partner base.

Sometimes people ask me, “Are you still committed to this partnership strategy?Are you going to go out and really scale up and hire all the people Oracle has or all the people IBM has to go after the enterprise?”And the answer to that is no.We’re going to hire a lot more than we’ve ever had in place before, particularly to go after accounts 2000 — or 2500 through 10,000, but we’re going to still leverage very much the kinds of companies on this slide to help us in the enterprise space.

Those are the challenges and competitors, the opportunities, the digital nervous system.What does all that mean in terms of what revenue might look like?I don’t know, I won’t make specific comments about revenue growth, but I do want to at least highlight a few of the things that I think you ought to think about.

This is a slide that I chose to entitle, “You assume and we worry.”Just that these are assumed, okay?Just assumed, assumed, assumed in everything that I read and see coming out of the financial community.I don’t think you can assume them, but I just want everybody to really dig deep in their hearts and understand how much you’ve already based the phenomenal, phenomenal market factors into your views of our company.

The number one thing is PC growth.We did this exercise of three-year planning back in February just to sort of gauge how big the company might be.We finished the financial plan and Bob and I looked at each other, and it felt like a waste of time.Why was it a waste of time?By far the thing that we are the most sensitive to is PC growth.And Bob (Herbold) and I had made up the PC growth number at the start of the exercise.So all that work by all those people — we raised the PC growth rate by 2 percent –we dropped it by 2 percent.And it wasn’t nearly as sensitive as you might think in some of the other factors.

People are assuming strong PC growth.I don’t think that’s a given.I think we expect to have reasonable growth, but as Greg (Maffei) will show you later, we’re highly correlated and PC growth has been slowing.

People assume we maintain Windows share.That may be a safe assumption for this fiscal year, but certainly as you look at all the NC pilots out there, we know we’re going to have to put a lot of energy on this.

People assume we maintain Office share despite the introduction of New Suites coming from Corel, Kona from IBM.I think that there’s a lot of risk in there.

People assume no precipitous shift — as I choose to say it — from new Office licenses to upgrades.When we sell a new Office license, we get several hundred dollars.When we sell an Office upgrade, we get fewer, like maybe half as many dollars as when we sell the original license.So the degree to which people are replacing machines and replacing software, where we get half as much roughly, per license as when the universe was buying new machines and new licenses.

Greg will show you the replacement rates on PCs.It’s a very scary proposition unless we can increase the frequency with which people upgrade.People basically assume that we’re going to make piracy progress in developing geographies and in small business.The U.S. is a high piracy country.People always say, No it’s not, no it’s not.There is high, high piracy rate in this country not only in the home, but in small business.And people just assume it’s not there and that we’ll be able to grow in small business.

People assume places like China and India represent huge opportunities.They are a huge source of growth.When people say PC growth’s going to be 18 percent or 15 percent or 20 percent or 8 percent, believe me, countries like China and India are pulling that up.And also believe me, piracy is still sky high, and unless we turn that corner, our revenue per PC must decline because of those high piracy areas.

Windows NT Server.

People, you know, have seen it grow on a trajectory like this.You ought to understand that our ability to continue to grow market share with Windows NT Server is dramatically less today than it was 12 months ago.And what we’ve been doing is benefiting from two things, market growth and share growth.You go to a country like Japan today, we’ve got 70 to 80 percent market share today of the PC server market — 70 to 80 percent.There’s no more share growth out there in the second largest country — second largest market for us in the world.Minimal share growth, let me say.And people have to understand that.

The NT server business — and I’m not talking about all of Back Office; I’m really talking about NT server here — it’s going to have to grow a lot more like market growth in servers than share growth.It’s more about market growth than share growth, and this is the first fiscal year for which I think that’s a fair thing for me to say.There are countries where there is still some share growth, but it’s absolutely trending down.

The second category of things I like to call “You wonder, we worry.”I think these are areas where I get a lot of questions in terms of what the potential is.The first is Windows NT Workstation.People want to know, when will you start — when will we start to see in the marketplace high penetration rates of Windows NT Workstation on new computers?And it’s a very good question.

I think that we have seen those rates ramp up pretty nicely.There’s still some upside absolutely.I don’t know if we’re going to capture it this year or next.We’re certainly forecasting in our thinking to capture some of that up side.But the big issue is, when will we see 60, 70, 80 percent of new corporate machines bundled with NT Workstation?

For the very largest accounts, I don’t know, maybe 15 to 25 percent of machines come with Windows NT Workstation today from the hardware manufacturer.In small businesses, 1 or 2 percent, and in the home, zero.So there’s a lot of up side, but I think we’re going to have to do a lot of good work to capture the up side, and I think many of you already expect that of us.

Memphis.

No, we haven’t set the price on Memphis yet.People have asked me that about 25 times today.And frankly, it is hard to gauge what the size of the Memphis opportunity is.It is certainly smaller than the Windows 95 opportunity.And — I think this is a very important point for people to understand — we had an incredible year in fiscal year 97 with Windows 95 upgrades.Now, some of you may scratch your head and say, no, no, that was fiscal year 96, Steve.We had an incredible year in ’97.

And so Memphis, if it’s a big-time success, might only keep us even with the kind of revenue success we’ve had with Win 95 upgrades in fiscal year 97.

SQL Server, Exchange, and Small Business Server.

I’ve already told you that we’re a relatively small percentage of the server applications business.People want to know how fast will that grow?Well, that’s a very good question.I think there’s more growth potential in terms of a percentage increase year over year, but on a much smaller base than we have with Windows NT Server.

When will we really seriously start to dent Oracle?The Exchange business, when will it really catch up with the size of the Notes business?These are slow-burn propositions.They’re not going to happen in one year or two years.

You all assume, I assume, because people are so bullish on the stock, you all must assume we do it in three or four years.But in the next year, I think we’re going to have to work very hard and we’re going to have to invest very aggressively in pursuing the developers and some additional support for accounts in order to get that business.

Office upgrades from large accounts.

That may be an opportunity, actually, this year versus what most of you expect.Office 97 did launch in fiscal year 97.We’ve sold a lot, but if we do a good job, we may have some opportunity, but it’s hard to plan for in fiscal year 98.

And the last topic is revenue from home PCs.If you look at the total amount of money spent on home computer software per PC in the United States per year, it’s certainly under $50.Not counting productivity software, it’s certainly under $50 per PC per year.And it’s not growing very fast.Most of the spending, whatever it is, coming out of homes on their PCs must be going into Internet or online because it’s sure not being spent on retail software.And so we have to wonder. But a lot of that potential is shifting to online, where everybody understands the revenue may be longer term in coming, when it comes.

In addition to these revenue issues and goals, we have a number of strategic goals which are not obviously generators of big revenue in the next 12 months.First is building Internet Explorer share.Don’t be confused, we’re happy with 30 or 35 percent, but not delighted.We won’t be delighted until we can glide share to 70, 75, 80 percent of that business.And we’ve got a lot of work to do with no positive revenue, but it’s absolute expenditure of money to drive IE share.

I talked about defending against NC pilots.We have to build developer enthusiasm for cost.Now, what does that mean?I’m not going to get into the detailed technical description, but it is very important for us to keep developers focused in on our platform.It has no revenue impact this year, but we’re going to have to spend money to really get after developers and tell them how to build and help them build next-generation applications.

And we have to defend against the Sun middleware operating system.Some of you may wonder what that is.That’s probably a reasonable question.Sun says they’re not in the desktop operating system really except for their Unix workstation business.But embedded deep there in the bowels of good old Java, they’re basically building their own middleware operating system that runs on Windows and other platforms.

Today this middleware operating system of theirs is not very complete, and really very buggy.And while we love Java very much here at Microsoft, we don’t love that middleware operating system.We’re helping people use Java to build applications that really target the most popular platform, which is Windows.

Everybody here knows Windows has increased its market share in the last couple of years.Cross-platform support in some senses is less relevant today than ever before.Windows on the desktop has increased market share versus Mac.Windows on the server has increased, Windows NT has increased market share versus Unix, Windows NT Workstation versus Unix for the workstation market has increased share.So in some senses, this cross-platform thing is less and less relevant.

But really letting people take advantage of the Java tools and really taking advantage of the Windows operating system, will mean we’re going to have to put a lot of energy, a lot of time and a lot of money into that, and that will — and that is a defense against this Sun middleware OS.

There’s one strategic opportunity, though, that’s sort of a flip side to that that came out of our detailed examination of Sun.We’ve never really focused on Sun as much as they’ve focused in on us, but since they’ve been hitting us over the head with this darn middleware operating system and Scott, yap, yap, yap, yap, yap.

(Laughter).

So a bunch of us really got busy looking at where our opportunities are in really focusing on Sun.And we found some great opportunities.They’re all focused on this new middleware operating system.Unix workstations, though, are 40, 45 percent of their business.That is a market that is declining.That is a market where we can really reach out with NT workstation and grab share, grab units.That’s actually one of the few nice opportunities.Yes, it’s only about a million units a year, but it’s a great opportunity to just grab something new at the desktop level.

The low-end and mid-range part of the Sun server line is basically non-competitive today versus PC servers.This has nothing to do with operating systems, but you just compare price performance of their low-end and mid-range line against what Compaq, Dell, HP can do, and it’s non-competitive.And we need to make sure the market really gets that message because that is another good opportunity.It may only be another 50,000, 60,000 servers, but 50,000, 60,000 servers is a big increase in a market where we’re only doing a million units a year.

Internet servers.

More and more companies are understanding the advantages, not only price performance, but also flexibility and tools, of building their Internet servers with NT as opposed to Unix and Sun boxes.We’ve had a lot of contact and a lot of focus with Sun VARs.And the Sun VAR community sees this picture on the horizon, and we found them very, very open-minded to the kinds of things we have to explain about PCs and NT.

Another great market for Sun has been the telecommunications industry.It’s been a market that’s been hard for us to break into because of the traditional strength of Unix.But whether you’re talking about switching applications, PBX applications, value-added voice services, more and more ISVs are really busy building those things on top of NT now.

And for that high end part of the Sun server line, the part that really has been growing nicely for them, we certainly see not only in our own work, but in our partnerships with Compaq, Tandem, Digital, Dell, and others a chance to really go after that higher end of the Sun server line.

So there’s a bit of an up side potentially here.It’s not huge, but Sun’s focus on that middleware operating system has certainly galvanized us on this opportunity.

We need to make some investments in order just to paint the picture that I’ve been describing.Let me talk a little bit about those.

Evangelism.

And when we talk about evangelism, we really talk about having an in-depth connection, both electronic and in person, with the best, most influential customers in the market, the enthusiast end users, the IT professionals and the developers.

In the U.S. alone, there’s about nine million of those people.A source of traditional strength for us has been our relationships.But we know we need, in some senses, to re-strengthen that muscle.That means more marketing money and more people applied to really touching and reaching out to that community.That’ll certainly grow faster, faster than we might have expected in fiscal year 98.

Small business.

In the small business area, we think we need to do even more to reach out to this value-added channel that serves small business.There’s about 150,000 to 200,000 small companies that provide IT services to small businesses in this country.We’ve started to reach out to them over the last 12 and 18 months, but we’re going to scale up our investment in that area.

The telecommunications industry.

We need to put more resources not only in galvanizing the ISVs, but also working with the key customers, whether those are equipment providers, like Octel, Lucent, Alcatel, or whether those are network operators, like BT, AT & T, Swedish Telecom, et cetera.And so we’re putting a lot more energy into that area.And we’ll spend money.

Growth geography.

We see good growth in China and India, Southeast Asia, Africa, Latin America.We will grow expenses in those countries also potentially faster than sales.

There are a few other investments that I want to highlight for you.The first is Microsoft.com.Microsoft.com, our Web site, is the fourth most popular site on the Internet.We get about 90 million hits per day, we get about 16-1/2 million views, page views per day.And now with our new profile in technology, we know we get about 900,000 unique visitors to Microsoft.com every day.

Managing that is not an inexpensive proposition.Now yes, we can start reducing maybe over time some other marketing expenses, but we have a data center today in Redmond, we’re starting data centers in Japan and London to serve those parts of the world.And just buying enough bandwidth, enough of the DS3s to serve up all of the pages that people want from us are expensive.

The servers, the data center operations — you can see we run entirely on NT.We’ve got about five SQL servers, 28 HTTP servers, three download servers, three FTP servers.That’s all reasonably inexpensive.But the networking, the connectivity and running those data centers, particularly in the remote geographies, certainly adds expense.

This is an incredible asset for us.This really allows us to reengineer our relationships with our customers.It’s certainly our target within the year to be able to get at all of those influential people that we want to evangelize and hit them with electronic communication and pointers to things on our Web site on a very regular basis.

But this is an area of investment, and unlike Netscape, we don’t sell advertising today on Microsoft.com, so there’s no cost recovery against this.That’s not true on the other three popular Web sites, the two big search guides and on Netscape.This is the first non-revenue — the top non-revenue-generating Web site on the Internet today, but it’s an incredibly key investment for us.

The second area where we’re going to invest money this year — or the next area is in what we call the application developer customer units.I talked about the real position that both Oracle and IBM with Notes have in the hearts and minds of developers.We need to put in more resource to really reach out to developers not just at the Windows level, where we’ve been very successful.But we need to reach out to people to build around SQL server, Exchange and Small Business server.

So we are forming a new what we call customer unit, with a clear focus on people who would build Back Office style application.We’re going to put over 600 people in this group, and their jobs will be to evangelize developers, to work with solution developers in a very, very strong way, and also try to reach out to line-of-business managers to help them understand what vertical solutions exist in health care, telecom, public safety, accounting, ERP, etc.You can see on the chart some of the areas in which we’re choosing to put this kind of vertical or horizontal focus.

The next area of significant investment is in enterprise accounts, particularly this account base 2,000 to about 8,000.We want to expand those relationships.We particularly want to expand our relationships with the bigger accounts to build these digital nervous systems.We think people who take that tack will be a source, will get greater value out of PCs and be a potential source of greater revenue for us, and so we need to invest.

By the end of this financial year, we will have over 7,000 people — 7,000 people — focused in on the enterprise.People keep asking me, “Do you have enough?”I don’t know.It’s expensive to pay 7,000 people.Providing pre-sales information, architectural design support, business application support, development skills transfer, licensing, account management, people who are involved on a support basis with helping with the deployment and with operational issue resolution.Seven thousand people.

This is not the Microsoft of four or five years ago.Yes, we need partners, but we are not standing here as a 100-pound weakling anymore.We have a core competence that we can use to leverage our partners’ work and leverage our customers’ work.But it doesn’t come cheaply.It’s an important investment that we’re making over the next 12 months.

One of the most significant aspects of that is the investment we’re making in our support operation.Our support operation provides technical account management, operational issue resolution and technical services.You can see on the slide some of the things we do.Perhaps the more important thing is, all of the things shown in yellow are services which we’ve added in the last 12 months and which we now need to make sure we have adequate head count to pay for in the ensuing 12 months.

But we can be there today for our customers, 24 hours a day, 7 days a week.We can provide them supportability reviews, preventive maintenance.If you have a problem of a critical nature that we can’t resolve within a set period of time, within basically four hours, we will put people on airplanes today and get them out and get them in the customer site, work the issue.It’s certainly for top management, more and more of my time — I’m not going to say it’s 80 percent or anything, but more and more of my time is in really working and making sure our support operation can provide this kind of hot-fix turnaround.

Within the last three weeks, I’ve had hot issues at a major retailer.Store down, we got it back up very, very quickly, but we can’t reproduce the bug, therefore we can’t make sure we’ve gotten rid of it.Very important issue and our support organization is fully galvanized.

Web site down, major company in the health industry.Boom, we had that thing back, turned around, up and running again in three hours.The Web site was back faster, but the bug had been fixed and resolved within three hours.

And all of these are issues which are coming up to me, to Deborah Willingham, to other very, very senior people in our organization so that people can really depend upon our products and move forward.

Here’s a little bit of a busy slide, but the top order here says, Are customers satisfied?The yellow shows Very Satisfied and the blue shows Satisfied.You see, we have 95 percent customer satisfaction, and we have, you know, almost 75 percent very satisfied customers using our enterprise support operation.

We dispatch an engineer on all problems now within two minutes of receiving a call.Less than 1 percent of the incidents that are highest severity actually require a fix, but we’re turning those fixes around, by and large, within two weeks.That means from the time the customer reports the incident to the time we can diagnose, make the code change and get it back out.In the meantime, of course, we have to help the customer with the work-around.We’ve made incredible progress in this area, but it hasn’t come cheap and it’s not going to come cheap as we look forward to the next 12 months.

One of the customers who have been an extensive user of our premier support services has been the Toronto Dominion Bank.They’ve done these online and alternate delivery banking services and been incredibly excited about what they’ve gotten from us in support.There should be a guy’s face here.Let’s click, though, and hear if we can hear anything without the face there.

MICROSOFT CUSTOMER FROM TORONTO DOMINION BANK:(From videotape.)
“Premier support over the last couple weeks has been instrumental in helping us stay on our time line.We’ve been working with the technical account manager since the beginning of the project.The knowledge that person has built up helped us immensely, and without their help, without the knowledge they’ve gained of our organization and our product, that solution couldn’t have been put in place as quickly as it was.”

MR. BALLMER:We have a dedicated technical account manager who sits here in Redmond, who travels regularly to Toronto, who supports the Toronto Dominion Bank, who navigates and guides through our support organization whenever they have issues that need resolution.It’s the kind of support infrastructure that I think people have come to expect in the largest enterprises and which I feel fairly fully that we’re capable of providing today.

If you look more broadly at product support, even outside the enterprise space, there’s a lot of good news that we saw in fiscal year 97 which will not be repeated in fiscal year 98.In fiscal year 97, support incidents per unit were down 29 percent.That is per unit sold, we were getting 70 percent as many calls and electronic services requests as we did in fiscal year 96.And that’s really a testimony to the quality of the products.It’s a testimony to Windows 95, it’s a testimony to Office 97, and it’s a testimony to NT 4.And it is great, great news.

But that is one of the reasons why sales and support expenses as a percentage of revenue have in fact been very steady.In this next year, we certainly have a great concern that we’ll see a change, not because the products are getting
“buggier”
or more complex again, but that the customers are now attempting more and more ambitious things with our products.

We’re lucky a lot of our customers are now turning to the Web.We get over 100,000 visitors per day just to our support site on the Web, and that certainly might mitigate some of this turnaround.But as customers try to do things that are more complicated, they may call less often, but when they do call, we wind up having to work with them for a longer period of time, so our incident lengths have actually started to expand.

And as we look towards fiscal year 98, we certainly see support expenses as a percentage of revenue rising on a — well, on a fairly significant percentage versus fiscal year 97.

We also have a very well budgeted but not fully thought-through plan yet to increase our expenditures supporting the channel that deliver services to small businesses.This is a channel today that almost certainly has to pay for support from us in a very, very expensive way relative to their businesses, and we are looking for ways to invest more on our nickel as opposed to the channel’s nickel in selling and supporting small businesses.

All of this will give rise to an increase, as Greg will detail, in product support expenses as a percentage of revenue.

In the long term, I love our company.I said that at the beginning and I really mean it.But I think there are a lot of challenges.Everybody fully recognizes the upside.Not too many people are really fully seeing the challenges in our business.And I hope — as people might have expected, I hope I helped dimensionalize some of the biggest challenges, particularly on the revenue side.

Bob, to some extent, and Greg are going to talk a lot about the expense side, but a lot of what happens and a lot of our possibilities and a lot of the forecast and a lot of the sensitivity really depend on revenue.And I’ll tell you; there is a lot of variability in my own view of what revenue might be next year.People ask, and of course we don’t say, but people ask, “What’s your forecast?”The truth of the matter is there’s a wide range in what I think is possible for next year.And certainly most of the stock valuation today is predicated not only on doing wonderfully next year, but also the year after and the year after and the year after that.

I hope people really take to heart some of the challenges we have just in the next 12 months, let alone over the next couple of years, as Paul and Pete had a chance to detail.

So with that, I’ll say thank you very much, and I’ll turn things over to Greg Maffei.