What Steve Ballmer needs to do to save Microsoft's mobile strategy

A wasted decade on Windows Mobile threatens Microsoft's relevance in the future of computing. It's time for Microsoft to kill the Kin.


It appears that Microsoft CEO Steve Ballmer has finally woken up and realized that Microsoft's laughable mobile position is more than a product failure but a potential loss of relevance in the computing world of the future, where desktop PCs are like TVs and the real action is in mobile devices of all stripes.

This week, the heads of Microsoft's mobile and entertainment (Windows Mobile, Zune, and Xbox) division announced their pending departures. It's not a moment too soon, given the widespread doubts that the long-sagging division's mobile and music fortunes would revive under the status quo. (The Xbox is doing fine.) Unfortunately, Ballmer says the departures had nothing to do with Microsoft's slide into mobile irrelevance and that business will continue as usual. That's suicidal.

With Apple's huge lead in mobile with the iPhone, iPod Touch, and iPad; Google's Android surge in smartphones and perhaps soon in slate-style tablets; and Research in Motion's seeming lock on messaging devices, people might wonder why Microsoft doesn't cut its losses and shift more emphasis to the company's cloud strategy, which Microsoft's execs hope will be its next-generation platform dominator à la Windows.

Ballmer should know why: Because mobile devices are where the client figure is, and a Microsoft without a strong mobile position means Microsoft loses any hope of owning the emerging technology ecosystem. At worst, Google would own it; at best, it would be a combination of Google, Apple, and Microsoft.

Microsoft has wasted a decade -- it's been that long since Windows Mobile (née CE) did anything that mattered to customers - with meaningless updates on an operating system that showed signs of innovation in 2000 but quickly became a confused mess of desktop wannabe functions by 2004. Complacency clearly set in as Palm frittered its future on endless reorganizations and RIM stayed happily in the mud of messaging. Then the iPhone showed up in 2007 and changed the mobile world. Google saw it and after a rough start started to deliver serious alternative. Both now outsell the establishment mobile OS that Windows Mobile had meant to be.

Steve, what did your company do? It wasted several years on the hapless Windows 6.5 and the moronic Kin - talk about throwing good money after bad. Now you have to turn around the mobile gap, and fast. Here's what you need to do - and you have only the rest of this year to do it.

1. Kill the crap - and do the rest right or not at all

Microsoft produces a lot of mediocre software that it slowly fixes over a half-dozen or more iterations -- adding more crap along the way. I've never understood this business strategy, but it worked for Windows, so it's now accepted at Redmond. Practically every product from Office to Dynamics is developed this way.

For Windows Mobile, the crap caught up to it, smothering the operating system in a pile of excrement that as of Windows Mobile 6.5 could not longer be disguised for what it was. The new Kin mobile platform is also full of crap, lacing some interesting ideas around social networking on a badly designed user interface running on crappy hardware. It's Windows Mobile all over again.

The successor to Windows Mobile -- Windows Phone 7 - can't contain crap. It needs to be a good OS, with a UI that works at all levels. The operating system needs to be elegant, simple, intentional, and consistent -- something Microsoft has never been good at.

Instead, Steve, your company's engineers confuse elegance with decoration, simplicity with obscurity, intentionality with brutishness, and consistency with -- well, they don't know that word. If you provide three ways to do the same thing, clutter up the screen with menus and dialog boxes and radio buttons, badger users with "helpful" alerts and confirmations, rely on multifinger keyboard shortcuts, and change navigation techniques across apps, you're dead.

Everyone cites Apple as the master of this game, but I don't believe for a minute that only Apple's designers can produce such quality. Apple is a standout, but it doesn't have a lock on good design. Amazon.com, Acer, Renault, and Nike are some top-of-mind examples.

Ironically, Microsoft has occasionally shown itself capable of design excellence: The original Excel was the first application to show that a graphical UI wasn't just for pretty pictures, and its success is what I believe made Windows possible. Before that, the original Word was a very well-designed application in a sea of ungainly competitors back in the DOS era.

The bottom line: If you can't do something at a level of excellence, pull it. Add it later, when it's truly ready, as a free update. Get out of the crapware and shovelware business - you won't believe the kind of loyalty you'll get if you deliver quality products without compromises. (Just ask Apple.) I'm not suggesting everything be perfect - often, there's no such thing - but customers can spot half-assed and rushed a mile away, and you already are too well known for those.

Your new mobile OS needs to run on hardware that oozes quality, fit and finish, and confident capability. You don't need to clone the iPhone to get that; the HTC Droid Incredible is another example, as are the top-end PCs and laptops from Acer, Hewlett-Packard, and Dell. Don't add a ton of perpherals and ports; instead, provide just one or two MicroUSB ports, the fewest buttons needed (without making any one button do unrelated things), a really good camera for still photography and videoconferencing (and remember it's not the megapixels that matter but the CCD quality, despite what the Gizmoids say), an amazing screen with excellent multitouch sensing, a memory card slot, flexible radios (CDMA and GSM 3G, Wi-Fi, Bluetooth not limited to headsets and keyboards, and maybe WiMax), and great audio in/out. A distinctive look that isn't distracing would be icing on that cake.

Furthermore, you should allow only two models: one with a physical keyboard, and one without. Memory and maybe a couple other noncritical features can vary within each model line, but don't create confusion. You don't need 15 different Windows Phone 7 devices of varying quality. You've already been burned going down that road. If you can't bring yourself to having just two models (and thus picking only HTC or Samsung as your hardware partner), enforce very tough requirements on the permissible variability. And don't let HTC add its own UI overlay; like Windows Mobile before it, Android needed the Sense UI to be usable, which, in essence, means they screwed up the OS. If you don't need it, you've just permitted an inconsistency in your platform's core.

Next Page: Get the Basics right

Find your next job with computerworld UK jobs