“Growth continued to moderate, particularly in December, and we exited the quarter with Azure constant currency growth in the mid-30s,” Amy Hood said, on Microsoft’s call with analysts. “From that, we expect Q3 growth to decelerate roughly four to five points in constant currency. FX impact in Azure is about one point more than at the segment level,” she added.
And just like that, the bump Microsoft’s shares enjoyed from what analysts described as a “better-than-feared” quarter was gone. By Wednesday morning, the stock was lower. Insult to injury was an hourslong bout of technical problems affecting Outlook, Teams and other services around the world. The company blamed a network configuration change, which was subsequently rolled back.
For those who might’ve missed it, Azure revenue growth was 38% during the most recent quarter, low by historical standards, but more resilient than anticipated. The guide was crucial, and although Hood didn’t appear to suggest the situation was deteriorating more rapidly than expected, markets were apparently inclined to sell what investors already knew — namely that Azure growth was set to decelerate further going forward.
Asked by Bernstein’s Mark Moerdler what part of the slowing in the cloud business is attributable to “people optimizing what they already bought” and what’s attributable to “macro factors impacting demand,” Satya Nadella conceded that the company’s own promise to help customers “do more with less” is “sort of the first place customers go to.”
Nadella continued: “I think a little bit of what has to happen is the cycle time — where the optimization cycle finishes, the projects start and then the projects ramp.”
Hood chimed in. “I think it’s quite hard to separate from a driver perspective how much is optimization versus macro,” she told Moerdler. “It’s all related when you start to say ‘What’s the best ROI I can get on every budget dollar I spend?’ and our job as a partner to these customers is to help them do that.”
“Perfect. Thank you very much,” Moerdler responded.
That exchange might’ve been “perfect” for Moerdler, but I’m not sure it satisfactorily addresses the core question, which is this: Is the majority of the sales growth slowdown in Microsoft’s high-growth business the result of an observably deteriorating macro environment or isn’t it? Every major tech company will surely face similar questions when they report, particularly given planned layoffs affecting tens of thousands of presumably capable people.
It’s obviously true that disentangling optimization and reduced spending is difficult. They’re both cost-cutting measures, after all, but you’d like to think the largest, most well-run tech companies on the planet have at least some visibility into what’s down to macro headwinds and what isn’t.
Of course, markets are fickle. Just as Microsoft’s knee-jerk earnings rally disappeared in an instant during the call, the swoon tied to the guide could likewise give way to a “could’ve been worse” trade as the day (and week) unfold. But the message from Hood was unmistakable.
“As you heard from Satya, we are seeing customers exercise caution in this environment and we saw results weaken through December,” she told analysts. “We saw moderated consumption growth in Azure and lower-than-expected growth in new business across the standalone Office 365, EMS and Windows commercial products that are sold outside the Microsoft 365 suite.”
Ultimately, I suppose she’s right to suggest (tacitly) that disentangling optimization from macro concerns is a somewhat fruitless exercise. Caution is caution. A slowdown is a slowdown.
Is this “bad news is bad news (recession), or is this “bad news is good news (Fed will let up on the brake)”?
I think it’s important to acknowledge that Cloud service growth is directly tied to business growth. Given the current macro environment, many businesses are already shelving planned low priority projects. Existing infrastructure is already in the budget, so migrating something you’re hosting on premises to the Cloud is actually a new expense, that will also be shelved. Businesses are now expecting a recession this year and planning accordingly. This means optimizing costs whether that’s on prem or in the Cloud, the name of the game will be operational efficiency. This means even businesses that are using Cloud services like Azure will not be looking to increase that spend but “optimize” it by evaluating whether or not the cost of the Cloud services is necessary. This may actually lead to cutting the spend on Cloud in ways that introduce more risk by reducing the resiliency of applications. Perhaps senior leadership will no longer see the value in a multi-tennant DR plan that only provides value if there is a catastrophic failure? This will probably lead to Cloud services experiencing negative growth until the economy turns around.
cdameworth, I think your comment is spot on.
I’d also like to add that Azure is generally not the go-to cloud provider for software startups, but more for larger companies already locked into the windows ecosystem. MSFT can buy a lot of great companies, but they’ve been piggybacking off of the windows lock in for their entire existence and have never been able to actually make and execute a shiny thing that is a must have across the industry.
I also think there will be a realization among companies that they can outsource some of their cloud operations. For example, instead of hiring an internal team, they can push their data to a third party which operates in the cloud and spits out a useable dashboard or service or whatever — I think this is also going to optimize spend.
We’re gone from the days where a consultant can spend 50k on AWS in a week doing load balance testing (Yes, I did that) 😉
Sorry for killing the discussion, came back to say Visual Studio Code is a feather in MSFT’s cap. It’s shiny and a must have.
Thanks for the compliment! Microsoft is positioned really well. Their OS is used on the majority of client machines around the world. The addition of WSL2 provides a level playing field with other *nix OS’s and provides an avenue to build with tools that are not Windows compatible. To your point, VS Code is quickly becoming THE editor of choice for anyone not doing Java. Azure provides some really nice DevOps tools which I think will make them more ideal for leaner IT teams in the next couple of years. Chat GPT will make them more competitive with Google in the near term but I see the long term advantage of that tooling coming out in the form of a dedicated product to replace customer service and help desks. I’ll be buying if the shares get discounted. 🙂
With effort, any cloud provider could algorithmically parse out server usage to examine dev vs testing vs production and break out changes in growth among them all and more accurately address these issues, and have real answers. MS probably does this, I don’t know, but I would if I were them. But I also likely would demure and pretend I don’t know for my reporting to the press
Seems like the management team is important, again!
Everyone looked like a genius over the last few years, but it’ll be interesting to see how many Ballmers the extended rally and low rates had been hiding. I don’t think Nadella is one of them, but can think of a few at some other high-flyers….
“You don’t know who’s swimming naked ‘til the tide goes out”
It will certainly be interesting to see over the next few months!
I agree with the comment about Azure getting companies who are heavily Windows focused given the pricing incentives they offer. You can tell that they are trying to be more hip and with it with their AKS offering and as another comment mentioned their Azure DevOps stuff, but I don’t think companies without a heavy Windows footprint would choose Azure unless a higher up started shouting about “multi-cloud.” They’d also be extremely unlikely to convert an existing AWS customer because that customer would likely already have tons of data in S3 and workloads built around that data.
In general I get confused about earnings growth. I would have thought 38% growth would be objectively incredible even if it misses whatever forecasts are in place.