It’s Decision Time for HoloLens (Premium)

With the U.S. Congress nixing a U.S. Army plan to waste another $400 million on HoloLens headsets, it’s time for Microsoft to figure out what’s next. Indeed, it may have already figured out a way forward, if this week’s layoffs memo this week from CEO Satya Nadella is any indication.

Microsoft, he wrote, was now “allocating both [its] capital and talent to areas of secular growth and long-term competitiveness for the company, while divesting in other areas.” This will include “changes to [Microsoft’s] hardware portfolio.

I initially wondered whether this was bad news for Surface. But Dina Bass, a well-connected Bloomberg reporter put me straight.

“Some of the changes [the Microsoft memo mentioned] relate to the HoloLens goggle and the Integrated Visual Augmentation System, or IVAS, being designed for the Army, said the people, who asked not to be identified because the plans are confidential,” she reported. “Workforce reductions began Wednesday in Microsoft’s mixed-reality group, which makes HoloLens and IVAS, said one of the people. The company declined to comment on the actions affecting the HoloLens group.”

As Ms. Bass added on Twitter, “The troubled HoloLens business … is kind of still in search of a business model that attracts enough customers, and [Microsoft] is running shorter and shorter on time to find it.”

Troubled is putting it mildly. After a rough initial start, Microsoft issued a HoloLens 2 in 2019 that solved the biggest problems with the initial model, most notably by dramatically improving its field of view. But even a well-designed HoloLens needs to find its niche, and HoloLens has rarely done so: the product makes sense in a handful of vertical markets only—I’m a fan of the vehicle designing possibilities, in particular, but there are killer remote help scenarios too—and that’s the problem. There’s no broad, widespread use case for HoloLens.

This explains, I think, why Microsoft betrayed its moral and ethical high ground—you will recall that this is the firm that withheld facial recognition capabilities from law enforcement—and latched onto the U.S. Army in a last-ditch effort to save the platform. But that effort was doomed from the beginning. And when the first hints arrived that this bulky, view-occluding headset might not be the best solution for people holding weapons and being shot at, it should have come as no surprise. Focusing on the Army probably set back HoloLens as a more general platform by years.

The timing of the Army deal is also interesting: Microsoft announced the contract, which might have been worth up to almost $22 billion over 10 years, just as its mammoth $10 billion Joint Enterprise Defense Infrastructure (JEDI) contract was collapsing after renewed scrutiny by The U.S. Department of Defense. By the time the JEDI contract was canceled, HoloLens was there to pick up the slack. Or would have been, had it made any sense at all.

I also wonder if Alex Kipman’s belated exit from Microsoft isn’t tied up in this drama. It’s astonishing that Kipman, who led Microsoft’s HoloLens efforts, lasted so long given the publicity around “Me Too” and Microsoft’s friendly, inclusive new look under Mr. Nadella. This bad actor’s deeds were silently ignored by Microsoft’s leadership during a time in which it couldn’t afford to let HoloLens stumble. It now pledges to do better.

The HoloLens defeat also came after a much less damaging defeat to the similar but less advanced Windows Mixed Reality (WMR) platform, and nothing explains how well that’s going better than a visit to the product line’s website, where they’re still promoting a year-old headset that is among the very few compatible hardware peripherals ever made. Microsoft has already experienced what Apple is right now experiencing and what Meta’s troubles revealed: virtual reality (VR)/mixed reality (MR) and augmented reality (AR) may be exciting in some ways, and they may find some niche use cases, but these are not mainstream technologies and they are very, very expensive to develop. And it’s not clear if there will be any payoff in the short term.

But back to Microsoft, which previously delayed HoloLens 2 to ensure that the second device was a meaningful upgrade and has more recently delayed a planned HoloLens 3 for the same reason: with the HoloLens team scattered to the winds—Kipman was only the highest-profile loss; many other people on this team, including key executives like Don Box, had already left, often to defect to Meta—and the Army contract in tatters, what can be done, if anything, to save this platform? And should Microsoft even bother?

Here, as with digital music and voice assistants, Microsoft is about to be overrun—in the public perception, not to mention by revenues and unit sales—by a familiar foe. Yes, Apple has struggled to bring its AR/MR products to market, but the consumer electronics giant is forging ahead. And if products like HomePod and Apple Watch are any guide, we’ve seen an interesting resilience under Tim Cook, with Apple seemingly willing to take both risks and time to get its products right. Put simply, when Apple “invents” this new market, products like HoloLens and WMR will be distant memories, even to those few who were aware of them in the first place.

That might not matter, I know: Apple targets consumers, and HoloLens, well, I guess HoloLens targets the commercial market again now that the Army has been forced to retreat. What’s left?

In this time of belt-tightening, I will argue “not much.” And that unless Microsoft plans a future in which a smaller HoloLens team licenses the technology to third parties and supplies just the software, as it does with Windows to PC makers, there’s no hope for any future. Indeed, even that is unlikely to succeed in any meaningful way. “Meaningful” is a moving target, one tied to its plans for the so-called Metaverse. Which, to be frank, seems fanciful in commercial settings.

Should Microsoft just kill HoloLens?

This is a tough one. But this is also the Microsoft that killed Windows Phone, a platform with tens of millions of users, and laid off 18,000 employees attached to the project. HoloLens would be a much smaller hit, and with its potential diminished, I guess I’ll argue that doing so makes sense.

But sometimes doing the right thing just feels wrong.