Microsoft: The Lost Decade and the Comeback Nobody Bet On

In July 2012, the magazine Vanity Fair published a long, devastating article with a title that stuck like a label on Microsoft’s back for years: “Microsoft’s Lost Decade.” The writer argued that the company that had once terrified all of technology had become its version of a declining department-store chain, a once-dominant giant that had grown fat, bureaucratic, and afraid, frozen in place while Apple, Google, and Amazon invented the future around it.

The numbers behind the insult were brutal. At the height of the dot-com bubble in 1999, Microsoft had been one of the most valuable companies in the world, worth on the order of half a trillion dollars. By 2012, its market value had roughly halved while the rest of technology soared. Its stock had gone essentially nowhere for more than a decade. That same year, Apple’s iPhone alone was generating more revenue than all of Microsoft combined. The company that had once been a synonym for technological dominance was now the cautionary tale told to every incumbent: this is what happens when you stop being hungry.

And here is the part that makes Microsoft so instructive. Through that entire “lost decade,” Microsoft never stopped making money. It was hugely profitable the whole time. Windows and Office printed cash year after year. The company paid out billions in dividends. By the conventional measures of a business, it was thriving. And yet its stock was dead, and its reputation was worse than dead, because the market had decided that Microsoft’s best days were behind it. This is the central puzzle of Microsoft, and resolving it teaches one of the deepest lessons in all of investing: the gap between a good business and a good stock, and what it actually takes to close it.

The World Before Microsoft

To understand Microsoft’s rise, fall, and resurrection, you have to remember a world in which software was not where the value was.

In the mid-1970s, the money and the glory in computing belonged to hardware, the great machines built by companies like IBM. Software, the instructions that made the machines do anything, was treated as an afterthought, often bundled in for free with the expensive hardware. The idea that software itself could be the product, sold separately and licensed over and over, was close to heretical. But whoever owned the layer that everything else had to run on, the operating system, would own the most valuable real estate in the digital world. Two young men saw that before almost anyone.

The Founders

Microsoft was founded in 1975 by Bill Gates and Paul Allen, two friends who believed, against the conventional wisdom of the hardware era, that software would matter more than the machines. Their founding vision was as audacious as it was specific: a computer on every desk and in every home, all of them running Microsoft software.

The masterstroke came in 1980. IBM, the giant of the computing world, needed an operating system for its new personal computer and came to Microsoft. Gates, in one of the most consequential business decisions of the twentieth century, supplied the operating system, MS-DOS, but crucially licensed it to IBM rather than selling it outright, retaining the right to license the same software to other manufacturers. When an army of competitors began making IBM-compatible PCs, every one of them needed Microsoft’s operating system. Microsoft collected a toll on almost every personal computer sold on Earth. Windows, layered on top through the late 1980s and 1990s, turned that toll into a near-monopoly. By the late 1990s Gates’s improbable vision had essentially come true, and Microsoft was one of the most powerful companies in history.

The trait that defined Gates-era Microsoft was a ferocious, take-no-prisoners competitiveness, a will to win that crushed rivals and dominated every market it entered. It built the empire. And, turned on the wrong target at the wrong moment, it would nearly fossilize the company.

The Near-Death Moment

Microsoft’s near-death is the strangest of any legendary company, because it was not a crisis of survival but a crisis of relevance, and it unfolded in slow motion across more than a decade.

It began at the peak. In 2000, two things happened. Bill Gates stepped down as chief executive, handing the company to his longtime lieutenant Steve Ballmer, a brilliant salesman and operator but not, by his own description, a technologist. And a federal court, at the climax of a long antitrust battle, ruled that Microsoft had abused its monopoly and ordered that the company be broken in two. The breakup was eventually overturned on appeal and settled, but the message was unmistakable: Microsoft’s dominance was so total, and so aggressively defended, that the government had tried to dismantle it. The company emerged legally intact but chastened, distracted, and, fatally, still organized entirely around protecting Windows.

That last fact was the trap. Microsoft’s overwhelming instinct, for the next decade and more, was to protect the Windows-and-Office franchise at all costs, and that instinct caused it to miss, or deliberately smother, nearly every important new wave in technology. It missed the shift to internet search, ceding the field to Google and producing the also-ran Bing. It missed digital music, producing the Zune to compete with the iPod years too late. It catastrophically missed mobile, treating the smartphone as a threat to Windows rather than the future of computing, so that when the iPhone and Android arrived, Microsoft had no answer and its later phone efforts failed. And in 2007 it shipped Windows Vista, a version of its flagship product so bloated and disliked that it became a punchline.

Internally, the culture had curdled into something that actively prevented reinvention: a famously brutal system of ranking employees against one another that turned colleagues into rivals, rewarded politics over collaboration, and made the safe, Windows-protecting choice the only rational career move. The company was a fortress organized to defend a castle the world was quietly abandoning. And all the while, it kept making enormous profits, which is precisely why the danger was so easy to ignore.

The Drawdown: Thirteen Years of Dead Money

For the investor, the lesson lives in the stock chart, and it is merciless.

If you had bought Microsoft at its peak around the end of 1999 and held, you would have waited, by most reckonings, until the mid-2010s, roughly thirteen to sixteen years, simply to get back to even on the share price. Through that entire stretch, the company grew its revenue and profits substantially. The business got bigger. The stock did not move.

This is the puzzle made concrete. The “dead money” was not caused by a failing business; it was caused by a collapsing valuation. At the bubble peak, Microsoft’s stock had traded at more than a hundred times its earnings, an absurd, unsustainable price baked with euphoria. The lost decade was, in large part, the long, grinding process of that insane multiple deflating back to a sane one, even as earnings rose underneath it. Investors who bought the great company at the insane price spent more than a decade being punished, not because they were wrong about the company, but because they had paid a price that assumed perfection forever. It is one of the cleanest illustrations in market history of a truth every investor must internalize: the quality of a business and the wisdom of buying its stock are two completely different questions, and the bridge between them is the price you pay.

What Everyone Got Wrong

Mistake #1: Mistaking a valuation problem for a business problem.
Reality: the lost-decade stock was dead largely because it had been absurdly overpriced in 1999, not because the underlying business was dying. Earnings rose the whole time. The multiple, not the company, was deflating.

Mistake #2: Writing Microsoft off as a permanent fossil.
Reality: the “Sears of technology” verdict felt obviously correct in 2012 and turned out to be one of the worst calls of the decade. The same enormous, cash-rich franchise that looked like a relic was the platform from which the greatest comeback in modern corporate history would launch.

Mistake #3: Believing dominance is permanent, then believing decline is permanent.
Reality: investors made the same mistake twice in opposite directions. They priced Microsoft as if its monopoly would last forever, then priced it as if its irrelevance would. Both extremes were wrong, and both were expensive.

The Inflection

The turn, when it came, did not come from a product. It came from a person, and specifically from a change of heart.

In February 2014, Microsoft appointed its third chief executive in its history: Satya Nadella, an insider who had quietly built the company’s cloud business. Nadella did something subtler and harder than launching a hit gadget. He changed what Microsoft was about. He retired the old obsession with defending Windows at all costs and declared the company “mobile-first, cloud-first,” which in practice meant a heresy: Microsoft would put its software, Office above all, on rival platforms like Apple’s iPhone and Google’s Android, meeting customers wherever they were instead of trying to force them onto Windows. He poured resources into Azure, the cloud-computing platform that let businesses rent Microsoft’s vast computing infrastructure. And, most importantly, he set about dismantling the brutal, political internal culture and replacing it with one built around learning, humility, and collaboration, a shift he described as moving from a company of know-it-alls to a company of learn-it-alls.

The results vindicated the bet completely. Office shifted from a boxed product sold once to Office 365, a subscription generating recurring revenue forever. Azure grew explosively, becoming one of the two dominant cloud platforms on Earth, its revenue eventually surpassing seventy-five billion dollars a year. The cloud and subscription engines transformed Microsoft from a company milking aging franchises into one of the central infrastructure providers of the modern, AI-driven economy. By 2021 Microsoft crossed a two-trillion-dollar valuation. In January 2024 it overtook Apple, however briefly, as the most valuable company in the world. And in 2025 it crossed four trillion dollars. The fossil had become, once again, the future.

The Moat

Microsoft’s moat, like Apple’s, is ecosystem lock-in, but of a different and arguably stickier kind, because its deepest roots are in the enterprise.

The world’s businesses run on Microsoft. Windows, Office, the Exchange and Active Directory systems that quietly power corporate IT, and now Azure and the Microsoft 365 cloud, are woven so deeply into how organizations operate that replacing them is a project measured in years and fortunes, not weeks. Employees know the tools. IT departments are built around them. Switching costs are not just high; for a large enterprise they are close to prohibitive. That lock-in, originally built around Windows, has been successfully extended into the cloud, which is why Nadella’s pivot worked: he carried the captive enterprise base from the old fortress into the new one.

On top of that sits the structural advantage of cloud computing itself, vast scale that lowers cost per unit and a customer base that, having built its operations atop Azure, finds leaving almost unthinkable. Microsoft took a moat built for the desktop era and rebuilt it, larger, for the cloud and AI era. That is the rarest feat in business: not defending an old moat, but successfully digging a new one before the old one runs dry.

The Wealth Created

Consider two investors, because Microsoft, more than almost any company, shows that when you buy matters as much as what you buy.

The first bought Microsoft at the dot-com peak around the end of 1999, paying a euphoric price. That investor endured the entire lost decade, more than thirteen years of dead money, the public humiliation of the “lost decade” headlines, and the constant sight of Apple and Google racing ahead, before finally getting back to even and only then beginning to profit from the Nadella resurrection.

The second investor bought during the despair, in the early 2010s, when Microsoft was written off as a fossil and traded at a modest multiple. That investor caught the entire comeback, from roughly a quarter-trillion-dollar company to one worth four trillion, one of the great mega-cap returns of the era, multiplying their money many times over in roughly a decade.

Same company. Wildly different outcomes. The difference was entirely the price paid and the sentiment at the moment of purchase. The greatest opportunity in Microsoft was available precisely when the consensus was most certain the company was finished. That is the recurring shape of every legendary company: the asymmetry hides inside the moment of maximum pessimism.

The Alternative Timeline

A counterfactual, clearly hypothetical.

Picture the world where Microsoft’s board, in 2014, plays it safe and appoints another CEO in the Ballmer mold, someone who promises to defend Windows, double down on mobile to fight Apple and Google head-on, and protect the existing empire.

Microsoft, in that timeline, keeps fighting the last war. It pours billions more into a doomed phone platform. It treats the cloud as a sideline that might cannibalize Windows licensing, and starves Azure. It keeps the brutal internal culture that punishes collaboration. And it slowly, profitably, declines, exactly as the “lost decade” thesis predicted, becoming a shrinking enterprise relic living off Office renewals while Amazon owns the cloud and the AI revolution is built on someone else’s infrastructure. The most valuable company of the mid-2020s is not Microsoft.

It did not happen that way, because a board chose a cloud builder over an empire defender, and that builder had the humility to put Office on the iPhone and the wisdom to change a culture rather than just a strategy. The lesson is that the hardest reinventions are not technical but human: the thing that had to change at Microsoft was not its code but its character, and only a leader who understood that could have turned a fossil back into the future.

Why This Matters to Investors

Microsoft fits the thesis with an unusual twist, because it lived the thesis twice. The first improbable idea, software on every desk, succeeded so completely it became a monopoly. Then came the stretch where failure, or at least permanent decline, looked inevitable: the lost decade, the missed waves, the “Sears of technology.” And then the point where success looked obvious again: the cloud-and-AI titan worth four trillion dollars. The opportunity, exactly as the thesis demands, existed only in the dark middle, when buying Microsoft meant buying a company the smartest commentators had pronounced a relic.

The reason to study Microsoft is that it isolates a lesson the other companies only imply: a wonderful business can be a terrible investment if you pay the wrong price, and a written-off business can be the opportunity of the decade if the market’s pessimism has overshot. The greatest opportunities almost always looked terrible before they looked inevitable, and Microsoft, in 2012, looked like a dinosaur that simply hadn’t lain down yet.

Lessons in Order of Depth

On the surface — the move

Carry your moat into the new era instead of defending the old one. Nadella took the captive enterprise base built around Windows and moved it into the cloud, rather than dying to protect the desktop. The trader’s analogue is adapting your edge as the market regime changes, instead of clinging to the setup that used to work.

Below the surface — the Money

Price is everything. The identical company made one investor wait thirteen years for nothing and made another many times their money, and the only difference was the price and sentiment at purchase. The investor’s discipline is to separate “is this a great company?” from “is this a great price?”, and to act most boldly when a great company is hated.

Below that — the Mind

Beware the comfort of profitability. Microsoft’s lost decade was so dangerous precisely because the company kept making money, which let it ignore its growing irrelevance. The trader’s parallel is the danger of a strategy that is still mildly profitable but quietly decaying; comfort is the enemy of the honest reassessment that survival requires.

At the deepest level — the question

Microsoft’s turnaround was a change of culture, not code. So the deepest question it poses is about the nature of decline and renewal: when something you have built stops working, do you defend it harder, or do you have the humility to change what you are? Microsoft spent over a decade defending Windows and slowly fossilizing. It was saved the moment a leader was willing to say that the old identity, the thing that had made the company great, was now the thing holding it back. The deepest edge is not a winning position; it is the self-honesty to recognize when your greatest strength has become your greatest weakness, and the humility to become something new before the world forces you to.

The Legendary Scorecard

Category	Score	Notes
Founder Vision	10 / 10	“A computer on every desk” — and a software-first world
Innovation	9 / 10	Missed a decade of waves, then led the cloud and AI era
Execution	10 / 10	The cloud pivot is among the best-executed in business history
Moat	10 / 10	Enterprise lock-in, carried from Windows into Azure
Capital Allocation	9 / 10	Some wasteful Ballmer-era bets; superb reinvestment since
Wealth Creation	10 / 10	From ~$250B to ~$4T in roughly a decade
Durability	9 / 10	Deeply entrenched; faces AI-platform and regulatory risk
Historical Importance	10 / 10	Put a computer on every desk; now powers the cloud and AI
Overall Legendary	9.6 / 10	The definitive corporate reinvention

Scores are an editorial verdict on the standard eight-category scale used across the Greatest Companies series. The overall is a judgment, not a weighted average.

Company Timeline

• 1975 — Founded by Bill Gates and Paul Allen
• 1980 — Licenses MS-DOS to IBM, keeping the rights to license it widely
• 1985 — Windows 1.0
• 1995 — Windows 95
• 2000 — Gates steps down as CEO; antitrust breakup ordered (later settled)
• 2007 — Windows Vista widely panned
• 2009–2012 — Market cap near lows; stock flat for a decade; “lost decade”
• 2014 — Satya Nadella becomes CEO; “mobile-first, cloud-first”
• 2021 — Crosses $2 trillion
• 2024 — Overtakes Apple as most valuable company (~$2.89T)
• 2025 — Crosses $4 trillion; Azure surpasses $75B/yr; FY revenue $281.7B

Key Numbers

Founded	1975 (Redmond, Washington)
Founders	Bill Gates, Paul Allen
The lost decade	Stock ~flat 2000 to mid-2010s; cap roughly halved
The turn	Satya Nadella, CEO from Feb 2014
Azure revenue	over $75B/yr (FY2025, +34%)
FY2025 revenue	$281.7B
Market value (2025)	crossed ~$4 trillion

Related Reading

More Greatest Companies

• Apple: Ninety Days From Death to the World’s Most Valuable Company (the other side of the 1997 deal, and a rival ecosystem)
• Nvidia: The 90% Drawdown That Built the AI Age (the chips powering the cloud Microsoft sells)
• Amazon: The 94% Crash and the Cost of Being Right (the company that beat Microsoft to the cloud)

Lesson Hubs

• Innovation Cycles (how incumbents miss waves, and how they catch the next one)
• Competitive Moats (enterprise lock-in, the stickiest moat of all)

Across the Library

• The Dot-Com Bubble (Market Mayhem — the mania that set up Microsoft’s lost decade)

This article is part of the Greatest Companies series, adapted from the book Greatest Companies, now available on Kindle. Explore the wider framework in The Complete Trader’s Edge.

Frequently Asked Questions

When was Microsoft founded and by whom?

Microsoft was founded in 1975 by Bill Gates and Paul Allen, on the bet that software, not hardware, would become the most valuable part of computing. Its early fortune came from licensing MS-DOS to IBM in 1980 while keeping the right to license it to other manufacturers.

What was Microsoft’s “lost decade”?

Roughly 2000 to the mid-2010s, when Microsoft’s stock went essentially nowhere and it missed major waves in search, music, mobile, and social. Notably, the company stayed highly profitable the whole time; the stock was dead largely because it had been wildly overpriced at the 1999 peak.

How did Microsoft recover?

Satya Nadella became CEO in 2014 and reoriented the company around “mobile-first, cloud-first.” He put Office on rival platforms, invested heavily in the Azure cloud, shifted Office to a subscription model, and reset a toxic internal culture. Microsoft crossed $2 trillion in 2021 and $4 trillion in 2025.

What is the investing lesson of Microsoft’s lost decade?

That a great business and a great stock are not the same thing. An investor who bought Microsoft at the euphoric 1999 price waited over a decade just to break even, while one who bought during the pessimism of the early 2010s caught one of the era’s great returns. The price you pay, and the prevailing sentiment, matter enormously.

What is Microsoft’s competitive moat?

Deep enterprise ecosystem lock-in. The world’s businesses run on Windows, Office, and Microsoft’s back-office systems, and increasingly on Azure and Microsoft 365. Switching away is so costly and disruptive that the lock-in, originally built around Windows, now anchors Microsoft’s cloud business too.