Yesterday, Microsoft announced that they were writing off $900M in Surface RT inventory. This is based on price reductions on Surface RT to clear inventory. If we assume that Microsoft factored in $150 per unit and we do some simple math, we can then estimate that Microsoft is sitting on 6M Surface RTs. This is an absolute abomination, and I don’t think this is a surprise to many that Surface RT didn’t sell well, but what is a surprise is the magnitude of the write-down. Even with nearly $1B in write-downs, I don’t think Microsoft will cancel Windows RT and I want to share my thinking.
I would be remiss if I didn’t first give my opinion on why Windows RT didn’t sell well. First, I disagree with the notion that it has to do with the dual tablet-PC nature of Windows 8, and for that matter, RT. Research I have conducted and research I have seen shows that once users actually use a use a touch-Windows device, they like it. It’s that trial that is the tough part. What doomed Surface RT, plain and simple, was the lack of premier apps and because the tablet market shifted to the 7-8″ form factor. This isn’t the main topic of this post but I needed to weigh in.
To better understand why Microsoft will keep investing in Windows RT, we have to know why they invested in it in the first place. When Microsoft would have had to make the decision to support an ARM-based Windows RT, Intel did not have a competitive mobile part and had just come off of some very public mobile failures, Menlow and Moorestown. The CloverTrail schedule was risky, too, and Microsoft felt that they needed lower power ARM-based SOCs to meet the battery life bar set by the iPad and the Motorola Xoom. The other factor is that in the minds of both Microsoft and Intel, any dollar invested by an OEM into each others products, is a dollar that they lose. Microsoft is interested in cheap hardware so they can charge more for software. Intel is interested in cheap software so they can charge more for hardware. Makes sense, right?
The first reason Microsoft will keep investing in Windows RT is to keep Intel competitive on tablets. Microsoft thinks that if they don’t hold something over Intel’s head, they won’t see solutions in the future as competitive as Bay Trail which, at least on paper, looks very competitive for holiday 2013 Windows 8-based tablets. Microsoft is also seeking to lower prices on 7-8″ tablets, and they see ARM-based SOCs from someone like Rockchip or Huawei providing that cost reduction necessary to enable Microsoft to charge more for software or lower the product street price. We also need to factor in phones. Windows Phone 9 will most likely share the same kernel as Windows RT (9) and therefore it would make sense to cease development now for ARM to revive it a few years later. Finally, Microsoft is thinking wearables and IoT devices based on this shared Windows RT (9) kernel, and so far, Intel doesn’t have a roadmap that would provide this level of performance/watt necessary to last weeks on a single charge.
So even with nearly $1B in “losses” racked up so far, Microsoft will trudge on, because they believe that they need ARM-based silicon to cover all their product segment bases and increase the price of their software to OEMs.