Amazon Web Services (AWS) may be the cloud leader, but Microsoft Azure is gaining fast and should surpass AWS revenues by the end of 2014 — at least, according to Nomura analyst Rick Sherlund.
Still, as well as Microsoft may be doing, Sherlund’s $5.7 billion revenue target for Azure doesn’t stand up to serious scrutiny.
Azure on fire
Let’s be clear: Microsoft Azure is doing well — really, really well.
Though neither Microsoft nor Amazon break out their cloud services revenues, Microsoft noted in its last earnings call that cloud revenue grew 147 percent year-over-year. This led Nomura analyst Sherlund to predict Microsoft will hold the crown as top cloud by revenue by the end of 2014.
While Sherlund’s contention is almost certainly wrong (more on that below), Microsoft has concrete Azure numbers, shared at WPC 2014, that should buoy even the most ardent Azure admirers. As Neowin’s Brad Sims captures:
57 percent of Fortune 500 companies now use Azure
More than 300,000 active websites
More than 30 trillion storage objects
3 million requests per second
13 billion authentications per week
More than 1 million SQL databases in Azure
300 million Azure Active Directory users
1.65 million developers registered with Visual Studio Online
All this is impressive. More impressive is Microsoft CFO Amy Hood telling investors on Microsoft’s earnings call that Azure storage revenue doubled and compute tripled over the past year. No wonder Microsoft has been investing so heavily in its data centers to the tune of $15 billion.
As big as these numbers may be, they don’t add up to the industry’s biggest cloud by any metric.
Cloud = Azure?
Not that Microsoft is guilty of plumping its numbers. While some have wanted to translate Microsoft’s "cloud revenue" to "Azure," Microsoft hasn’t crossed this line. As Hood said in the earnings call:
Our commercial cloud revenue grew 147 percent this quarter, driven by both Office 365 and Azure. Our commercial cloud annual revenue run rate more than doubled this year and now exceeds $4.4 billion and with this rapidly growing scale we continue to expand our cloud gross margins.
That’s awesome. It’s big — but it’s not Azure. In fact, it’s not even clear how much of this revenue derives from Azure — or whether it’s true cloud revenue at all.
After all, it would seem that the overall cloud revenue number includes Services Provider License Agreement (SPLA) income — that is, rented software to Microsoft services partners, much of which may simply be part of a volume contract that may represent software sold but not ultimately used.
This is reminiscent of Microsoft in 2013. Back then, Barb Darrow dissected Microsoft’s "Azure is a $1 billion business" and found the $1 billion appeared to include all sorts of noncloud revenue, like "software provided to partners," which "probably means Microsoft is lumping in sales of on-premises software."
Amazon: Still on top
Which brings us back to Amazon: While AWS acknowledged slowed growth during its last earnings call, it also revealed that AWS usage climbed 90 percent over the past year.
Given that AWS offered five times the utilized compute capacity of the other 14 cloud providers in the Gartner Magic Quadrant combined as of August 2013, 90 percent growth is pretty impressive.
Guess what? So is Microsoft’s.
Maybe, just maybe, it will become even harder to separate Azure growth from the rest of Microsoft’s cloud business. After all, as Microsoft Technical Fellow Mark Russinovich told InfoWorld, "all roads lead to Azure" within Microsoft, "So there are parts of Office 365 that actually run on Azure today. The goal is to eventually have everything running on Azure."
Regardless, however much of a lead AWS has today, Microsoft’s hybrid strategy has real legs. How well Azure is doing may be in question but why Azure is doing well is not. As Gartner analyst Lydia Leong calls out, "Microsoft’s comprehensive hybrid story, which spans applications and platforms as well as infrastructure, is highly attractive to many companies, drawing them towards the cloud in general."
Microsoft, in short, won’t have the biggest cloud by the end of 2014 — not even close. But it will almost certainly continue to build out public cloud services that enterprises buy because they help carry their data center assets forward into a cloudy future.