|
Steve Ballmer CEO MICROSOFT on the Cloud
|
Global CIO: Steve Ballmer Interview: 'Hockey Stick' Cloud Growth Ahead
Microsoft CEO talks about new competition with Google, Amazon, and Salesforce, and why CIOs now are ready for cloud computing.
By Chris Murphy, InformationWeek
April 20, 2010
URL: http://www.informationweek.com/story/showArticle.jhtml?articleID=224400787
Microsoft CEO Steve Ballmer has sent a message of late that Microsoft is "all in" when it comes to cloud computing. In an interview with InformationWeek editors, Ballmer made clear that this all-in cloud computing bet isn't merely a long-term, over-the-horizon play. The cloud growth--"hockey stick" growth, he said -- is taking off right now. And Ballmer makes a passionate case for how the investment Microsoft has made in cloud computing products and infrastructure over the last five years makes it different from rivals Google, Amazon, and Salesforce.
We pushed to clarify--when does that hockey stick growth take off? We're not there yet, right? "I don't know. It sure feels like we're there today to me," Ballmer said.
He added, however, that most lines of business software--industry-specific applications or transaction systems, for example--aren't going to the cloud en masse yet. Platform as a service offerings, like Microsoft's Azure, haven't taken off. But with what he calls "information worker infrastructure" -- think Exchange, SharePoint, and Office software -- CIOs are ready to move quickly to the cloud.
"Look, I don't want to oversell or undersell, but the truth of the matter is there is not an enterprise customer I visit today where this is not an issue -- just not," Ballmer said. Any CIO considering an upgrade to the company's e-mail or other collaboration platforms has to at least consider going to a cloud-based infrastructure. "Everybody is saying, look, next time I touch anything, I'm going. If I'm not touching anything, maybe I don't go," Ballmer said. "But if I'm really going to touch something, I'm going to have this [cloud] discussion."
Ballmer kicked off our discussion in fine form, busting my colleague Fritz Nelson's chops for using an iPad hooked to a flimsy-looking keyboard to take notes. "Spend enough time and money, and you can make anything work like a PC," he cracked.
Text can never do justice to the intensity of Ballmer, who's rapid fire from topic to topic. So rather than present a straight Q&A, what follows is a series of extended excerpts from Ballmer's conversation with InformationWeek's Nelson, Rob Preston, and me, set up with the context of our conversation.
On where we are in terms of cloud adoption:
Ballmer: We have hit a point where things have been, let me say, moving along, incubating, coming along, being used, but I think we're really at a point where we'll see a transformational kind of hockey stick in the pace with which we will accelerate our efforts and our customers, most importantly, will accelerate their adoption, which is kind of the reason we get out and talk about things, and sort of put the stake in the ground at this stage.
We asked how the shift to cloud computing compares with the "Internet tidal wave" that Bill Gates predicted in a company memo in 1995. The cloud brings more business model change, as Microsoft adds huge new costs to run some of the world's largest data centers. Ballmer discussed the business model and competitive changes of cloud computing:
Ballmer: Operational process changes, pricing, and business model changes, for sure. I think there's huge opportunity, upside opportunity, for us from a profit perspective. It turns out that if you made $50 and 100% of it was profit, and now you take in $100 of revenue, and you have $40 of COGS -- these aren't real numbers -- it turns out your gross margin doesn't look as good in the second case, but your shareholders actually have more profit in the second case.
So, I say we've got a lot of opportunity. We actually have an opportunity to improve total gross margin dollars, total profit dollars. We've got a lot of competition, we've got a lot of issues, a lot of this, a lot of that, but we have opportunities, because where we used to do this much for our customers, we're now putting out value propositions where we do a lot more for our customers. And we can save our customers money we couldn't save them before, and that gives us an opportunity to make more money. On the other hand, we're going to have different competitive dynamics. If the competitor of the last wave, the primary competitors were Linux and Open Office, and none of that goes away, but I would say the competitors will be somewhat different. It will be Amazon. VMware is kind of a funny tweener I would say on this, because they don't really operate a service themselves. It will be Google, Salesforce, etc.
So, business model changes, operational process changes, and technology changes, but we're already someplace between five to 10 years into some of those changes, for primarily the technology shifts.
Azure is Microsoft's cloud computing platform. Microsoft's betting companies will use Azure to develop applications differently to take advantage of cloud computing's flexibility, and that they'll move those apps between in-house data centers and Microsoft's cloud. On Azure:
Ballmer: The truth of the matter is there's nobody with an offer like ours in the market today, not even close. We're actually trying to help people do what they really will need to do for the modern time. You don't get that out of what Amazon is doing. Amazon is a great company doing great stuff, but they basically say give us your VMs and you can still muck around at the low level.
Well, it basically says let's not change anything. Let's deal with all the complexity of today and put it in the cloud. And they've done a very nice job, that's not accusatory. Different target.
Look, nobody builds cloud services like that, right? The good cloud services don't look anything like an enterprise data center. They're not managed that way. You cannot manage the scale of transaction, the scale of change, the geo distribution of large-scale services, you just can't manage them, either server by server or even VM by VM. That's not how we manage Bing, that's not how Facebook manages their stuff, it's not how Google manages their stuff. You manage the stuff, and in order to manage them in a completely different way, you build it in a somewhat different way.
... And if you look at it today, there is no offer like Azure from anybody else in the market. I mean, if you just go through it. Amazon has done a very nice job helping you take existing things and put them in the cloud. They've done a great job of that. One of the things we'd better add to Azure here sometime is the ability to just take existing things and move them to the cloud. That wouldn't have been a forward-looking thing for us to do as the owner of a development platform, but Amazon has done a nice job, and I give them credit.
But once you get past that, VMware has no service. They'll tell you, you can kind of roll your own and blah, blah, blah, blah.
You know, Salesforce is not a general-purpose programming platform. They're not trying to be. They've done some nice work, but they're not a general-purpose programming platform for large-scale deployment.
Google is kind of its own weird, funny proprietary environment, which I don't think has any hope of ever coming back and being run except out of a Google data center. We have to have a story and an approach where people can run our cloud in their data center. We call that the private cloud. Depending on who you're talking about, that's either viewed as very smart and forward-looking or very troglodyte and backward looking.
We talked about what is it that has CIOs excited about the cloud now:
Ballmer: Between let's say the first part of '07 and the end of '08, it just flipped. Conventional wisdom became I'm going to the cloud, from I hear you about the cloud, until today where people--it's really just a question of where are we. There's a mix of customers and a mix of application sets. It's not like people are saying I've got to take every bespoke application in my shop and move it to the cloud, but people are saying I have a set of bespoke applications I want to move to the cloud, because they have customer contacts, they have high variability in the demand, people are saying, God, I've got to go there for I guess what people like to call SaaS as opposed to PaaS. I don't know if people use these expressions.
But certainly as we think about the desktop, people would love the desktop to be a service in all of its richness, Windows and Windows application execution, collaboration, communication, productivity. People are pushing on that, as they are in CRM. SharePoint is kind of a funny one, and people say, yeah, I'd like it in the cloud, but the truth of the matter is when you use SharePoint as a front-end for enterprise data, the enterprise data set doesn't move to the cloud, people still want SharePoint on-premise for that purpose, and then they start saying, well, how do I connect my desktop environment in the cloud to enterprise data that lives inside the enterprise, the directory and security infrastructure. We're getting into I'd say the rub of the matter to make that stuff happen. We've done really well winning enterprise customers, particularly with the BPOS [Business Productivity Online Standard Suite, which includes Exchange and SharePoint] line of products, very, very well, but we have some losses, some of which have been well publicized. But what we find is the first thing anybody does, even when we lose, they get to work with us hard core on their Active Directory infrastructure, because the importance of a clear and well managed directory and security environment is so much more, because there's always some things you've got to bridge between what's going on in the cloud and what's going on on premise as you move data back and forth.
So, customers are saying it's a different world. You know, there's always a class of CIO who wants to be the avant-garde of avant-garde, as there would be in any other community, but there is in the CIO community, and they're really pushing out there even faster. But even I'll call it the -- I won't call it the lagging edge, but the mainstream people are saying I'm ready for a lot of things to move to the cloud. And we see it in the volumes, the customer signups. We absolutely see it.
When we asked what besides cloud computing is high on the CIO agenda, Ballmer cited collaboration:
Ballmer: You really do hear a lot about collaboration. You really do. And everybody means something a little different about what is collaboration. To some it's how do we get the equivalent of social networking in the enterprise, to some it's frankly SharePoint --it has kind of helped define the category for a lot of enterprise CIOs, and I think you all know that's been kind of a rocket ship in terms of its adoption. And it's a Swiss Army Knife of a sort -- it's an enterprise search product, it's a BI product, it's a traditional collaboration product, it's a rapid application development platform for certain kinds of workflow applications. It's a lot of things. But it's about bringing people together in some kind of I'll call it semi-structured way, and that's what I think people mean by collaboration.
That is very much on people's minds. Certainly what they're doing with voice and presence and networking and IM is on people's minds. I put that all in this one topic.
Ballmer noted collaboration was hard to get on the CIO agenda five or six years ago. So we asked what had changed:
Ballmer: I think the theory of the situation is the push from the users -- you remember a lot of IT is end user driven -- the push from the users has continued to accelerate. That's one thing.
And No. 2, I think most people -- many people, not most -- many people have done their ERP thing. Let me just say it that way. They've done their ERP thing, and whatever the level of investment was that has gone into it is ramped down. It doesn't mean it's gone away. There was a period of rapid acceleration in ERP. The same thing on the Web site--got to have a Web site.
So, there were two big sort of phenomena if you go, say, probably from '95 to about 2005. Those two things sucked a lot of let's call it brainpower and capacity from the IT crowds in most companies. And I think both of them -- it's not like either one of them has gone away, but they're not ramping anymore, and the push from the users continues to ramp for better and better collaboration infrastructure as they see more and more things in their personal lives.
Recommended Reading:
|
