In November 2023, Broadcom acquired VMware for $61 billion, the largest software acquisition in history. Within a few months, the entire business model changed. Perpetual licenses were phased out in favor of subscriptions. Products were consolidated into a few offerings. A minimum purchase requirement of 72 cores is imposed, and a 20 percent penalty is applied to late renewals. And the bills are going up.

And not just a little. In Europe, price increases of up to 1,500 percent have been documented for some customers. A January 2026 survey of large North American companies illustrates the magnitude of the shock: nearly six in ten saw their prices rise by more than a quarter, and more than eight in ten are now reducing their use of VMware. The same report notes that the majority of organizations have changed their strategy twice or more in two years. It’s hard to unplug what everything revolves around.

What Broadcom Revealed—and Didn’t Break

The instinct is to put Broadcom on trial. Tempting, but pointless. Broadcom hasn’t broken anything. The company is extracting value from an asset it paid dearly for, with formidable efficiency: its operating margins on infrastructure software reach 77 percent. For a shareholder, that’s a success. For a customer, it’s a lesson in dependency.

What the acquisition revealed is a truth we’d rather not face. An entire industry had built its foundation on a single supplier, assuming that yesterday’s conditions would last forever. VMware virtualization was everywhere—perpetual, invisible, taken for granted. The day ownership changed, the lock tightened a notch, and everyone discovered they didn’t have the key.

The restrictions didn’t stop at price. Broadcom sent cease-and-desist letters to users without subscriptions and filed lawsuits against major clients, including Siemens’ U.S. operations. This is the disruption in its commercial and legal form. Not a server you shut down remotely, but a relationship that’s tightened until staying costs more than everything else.

The Wrong Reflex: Switching Providers

Faced with the bill, many made the familiar, reassuring choice. They fled VMware for another proprietary vendor, with Nutanix leading the way. The migration sells well, the competitor welcomes them with open arms, and they breathe a sigh of relief. They’ve just switched locks. Same trap, different logo, and in three years, the same conversation when the new owner decides, in turn, to squeeze more value out of the asset.

Sovereignty isn’t gained by replacing one locked-in vendor with another. It’s gained by asking the only question that matters: Can we leave? What is the actual reversibility, at what cost, in how much time, and to what alternative? This question has concrete answers. Open platforms like Proxmox or OpenStack don’t eliminate dependency; they make it reversible, because the format remains readable and no single party holds all the power. Others have chosen to stay with VMware by negotiating price caps and contractual exit rights, which is defensible as long as the decision was made with full knowledge of the facts.

The Lesson, Beyond Virtualization

The VMware case is by no means unique. It is a real-world demonstration of what proprietary lock-in means—and of the cost of a lack of reversibility. The technology was excellent, the dependency comfortable, and it was precisely that comfort that made the bill so steep the day the balance of power shifted.

The question to ask didn’t arise with the acquisition. It should have been asked before: what does my business rely on to run, and what would it cost to break free from it? A single supplier that works well is still just a single supplier. As long as you can switch to another one within a reasonable timeframe, you’re in control. The day you can no longer do so, you’re a tenant, and the landlord sets the rent.

Sources