Preston Gralla: Memo to Nadella: Copy IBM

Can a company that's seen as a technology dinosaur re-establish itself as a powerful and profitable leader on the cutting edge? It has happened before, and if Microsoft CEO Satya Nadella wants a primer on how it's done, he need only look at Microsoft's onetime partner IBM.

Much like Microsoft, IBM spent years at the top, acclaimed as an innovative and seemingly all-powerful company. Both were at one time the default vendor for a lot of IT departments. But by the late 1980s (about the time that Microsoft was really coming into its own), IBM had become a bloated and slow-moving behemoth, living off the glory days of the mainframe and tied to low-margin businesses. The PC was ascendant, and though IBM had had much to do with that, it was having a hard time cashing in amid cutthroat competition. Worse, the new client/server architecture was seriously undermining its mainframe business. IBM flailed as it tried to set things right, at one point organizing itself into business units that operated somewhat like separate fiefdoms.

Earnings, which had consistently been above $5 billion in the early 1980s, fell to $3 billion by 1989. In 1992, IBM had what was then the biggest one-year loss in U.S. history, a whopping $8.1 billion.

The following year, IBM had a new CEO, Louis Gerstner, who undertook a radical makeover. The company abandoned commodity, low-margin products such as hard drives and personal printers and eventually sold off its PC business to Lenovo. It invested in middleware and targeted high-margin businesses, refashioning itself as a global services company and technology integrator.

IBM's labs thrived as well. They helped transform the company's image, from doddering has-been to an acknowledged leader in artificial intelligence and data analytics. Some high-profile publicity stunts helped, but only because the technology behind them was solid. In 1997, IBM's Deep Blue computer beat world chess champion Garry Kasparov in a six-game match. In 2011, its Watson AI machine dominated on Jeopardy! as it beat two previous champs. Today, Watson technology is being commercialized, making decisions about lung cancer treatments at Memorial Sloan Kettering Cancer Center.

Microsoft's situation in 2014 is eerily similar to IBM's in the late 1980s. Windows is its mainframe, undermined by newer technologies that it hasn't been able to do much with, beyond offering a few also-ran products. Chief among these is mobile, which is giving companies like Google and Apple a side door into enterprises that once would never have looked beyond Microsoft.

So what can Nadella learn from IBM? Plenty. Most important is to follow IBM in becoming a services company while turning away from what had been its greatest strength -- hardware in IBM's case and software in Microsoft's. Nadella has started doing this, by allowing Nokia to sell Android phones that carry Microsoft services, for example, and by giving away Windows to manufacturers of devices under 9 inches, because those devices will carry Microsoft services.

But Nadella still hasn't completely abandoned the business strategy he inherited from Steve Ballmer, which included devices as well as services. Nadella should take a page from IBM and abandon hardware as well -- it's a low-margin business, and Microsoft can't compete there. The company has already racked up $1.2 billion in losses on its Surface line, and more losses are certainly on the way.

If Nadella follows IBM's lead, it won't take 20 years to restore Microsoft's lost luster. Within five, it could again be in the forefront of the technology industry, not trying to play catch-up.

Preston Gralla is a contributing editor and the author of more than 35 books, including How the Internet Works (Que, 2006).

Read more about management in Computerworld's Management Topic Center.