Broadcom took VMware. Now what?

Roger: A university in the UK went public with their number. Their annual VMware support cost went from about forty thousand pounds to half a million. Twelve and a half times, for the same virtualization estate, because the acquirer rewrote the catalog.

Laura: And the scary part is not the number. It is that the playbook worked, and every acquirer in software watched it work.

Bella: Today, Broadcom and VMware, the most consequential repricing in enterprise software, and what it teaches about the next acquisition of a tool you depend on. This is Off List. Read the paper before you sign it.

Bella: Welcome back to Off List. Bella, with Roger, Bill, and Laura. Today is a case study episode, and the case is the one the whole industry is still absorbing: Broadcom's acquisition of VMware, and what happened to every VMware customer's bill afterward. We are going to walk the timeline of what actually changed, the honest options a customer has now, how to find leverage when the vendor has correctly concluded you cannot leave, and then the bigger lesson, because this was not a one off. It was a demonstration. Bill, you run VMware. Start with the day the world changed.

Bill: Late twenty twenty three, the deal closes, and within weeks the new owner does four things at once, and the speed was the message. Perpetual licenses, gone, you can no longer buy VMware, only subscribe to it. The product catalog, collapsed, dozens of products bundled into a small number of large suites, so the thing you actually use is now only sold inside a package of things you may not. The channel, purged, thousands of partners cut, which for mid sized customers meant the person who managed your renewal for a decade simply no longer could. And the prices, well, the prices are the episode.

Roger: And the strategic clarity deserves respect, coldly speaking. Broadcom said out loud what they were doing: focus on the largest customers, move everything to subscription, monetize the installed base. They were not confused, and they were not embarrassed. Every customer was rebalanced from partner to product, from perpetual to rental, in one motion. The only question on the buyer side was, and remains, how much of it lands on you.

Bella: The prices. Give people the real shape of the increases, because the range is so wide that averages hide it.

Roger: The honest range, from the reporting and from the deals we have seen: transitions from perpetual and standalone licensing to the new subscription bundles have landed anywhere from one hundred fifty percent increases at the merciful end to over a thousand percent at the other. Small businesses commonly report three hundred fifty to four hundred fifty percent. The public reference case is that UK university, forty thousand pounds of annual support becoming half a million under the new bundle mandate, twelve and a half times. And the reason the range is so wide is that the increase is not one change, it is three stacked: subscription pricing replacing support on licenses you owned, the bundle forcing you to pay for suite components you never used, and the core mathematics.

Bella: Do the core mathematics, because this is the part that quietly doubled bills a second time.

Roger: VMware pricing is per processor core, and there is a minimum. When the new model launched, the minimum was sixteen cores per processor. In April twenty twenty five, that minimum jumped from sixteen to seventy two cores per processor. Seventy two. If your server has a ten core processor, you pay for seventy two. A modest two processor host with sixteen core chips, thirty two real cores, licenses as one hundred forty four. For small and mid sized estates running efficient hardware, that single change repriced everything again, a year after the first repricing, and it specifically punishes the customers with the least leverage.

Bill: And the small print kept coming. There is now a late renewal surcharge, miss your subscription anniversary and the new term carries a twenty percent penalty on the first year. Which, for a company like mine with a renewal calendar, is an annoyance. For the mid sized company with no sourcing function, the exact customer already hit hardest by the core minimum, it is a tax on being under resourced. Every mechanism in this catalog change lands heaviest on whoever can defend themselves least.

Laura: Which is the pattern to name before we get to options: none of this is a negotiation that went badly. There was no negotiation. It is a catalog. The lesson of the whole case is what happens when a vendor concludes it does not need to negotiate with most of its customers, and the defense has to be built before that conclusion gets made about you.

Bella: So you are a VMware customer holding one of these quotes. The four options, honestly costed. Roger.

Roger: Option one, negotiate and stay. Real, for some. Broadcom does discount, meaningfully, for large customers on long commitments, three to five year terms. If you are big enough to matter to their focus list, a multi year commitment with a hard price cap can pull the increase down substantially, and for a genuinely VMware dependent estate that can be the right trade. The cost is the lock in, you are signing years of subscription with an owner who has demonstrated exactly how it treats its installed base. Get the caps in writing, on renewal too, because you have seen what the catalog does between terms.

Bill: Option two, migrate. The alternatives are real now in a way they were not five years ago. The hyperscalers for what belongs in cloud anyway, the other on premise hypervisor platforms, the open source stack for the adventurous, and the market has matured specifically because Broadcom created a refugee population. But cost it honestly: a virtualization migration is one of the most disruptive projects an infrastructure team can run, it is measured in years for a large estate, and the tooling around the hypervisor, backup, monitoring, automation, all of it has tendrils. The migration is real leverage precisely because it is really painful, and vendors can tell the difference between a company that priced the pain and a company waving a brochure.

Laura: Option three, the partial exit, and this is the one I push because it is available to almost everyone. You do not migrate the estate, you migrate a slice, the new workloads, a branch environment, the dev and test tier. It caps the vendor's growth into your future, it builds real migration muscle on low stakes workloads, and it converts your walkaway from a brochure into a demonstrated capability. Level two leverage, from the mistakes episode, built deliberately.

Roger: And option four, for perpetual license holders specifically: stand still. You own those licenses. Third party support exists for keeping an owned VMware estate running without a subscription, at a fraction of the cost, and it buys you years of runway to decide calmly instead of renewing under the gun. It is a bridge, not a destination, the estate ages, but a three year bridge you control beats a five year subscription you resent. The vendors hate this option, which tells you it works.

Bella: The bigger lesson. Because the reason this case matters to people who never touched VMware is that it worked. The model was proven at scale, in public, and every acquirer and private equity firm in software has the case study. Bill, what does the sourcing discipline look like after Broadcom?

Bill: It looks like treating acquisition risk as a standing category of vendor risk, the same as security risk or viability risk, reviewed on the same calendar. Concretely, three habits. First, concentration awareness: know which of your load bearing vendors have the profile, dominant in a niche, sticky installed base, flat growth, the exact shape that makes monetize the base the obvious play for a buyer. That profile is scoreable, and your most embedded vendors should be scored. Second, contract survivability: the clauses that matter in an acquisition are the ones almost nobody negotiates. Price caps that survive assignment. Renewal terms that bind successors. Termination rights on change of control, even just for convenience with notice. The day the deal closes is too late to ask, and the decade before, those clauses are cheap because nobody prices them.

Roger: I will second the clause point with the field version: since the VMware case, I put change of control language in every significant negotiation, and vendors have mostly stopped resisting it, because their own sales team knows exactly why buyers are asking. The market repriced that clause from paranoia to prudence in about eighteen months. Use the window.

Laura: Third habit, and it is the cultural one: keep your exits warm. The Broadcom case punished monocultures. The companies that absorbed it best were the ones already running some workloads elsewhere, already benchmarking alternatives annually, already treating portability as an architecture requirement instead of a procurement fantasy. You cannot make a load bearing vendor replaceable overnight. You can keep the replacement muscle from atrophying, and the vendors can tell which kind of shop you are, and it prices.

Bella: The signals that the next one is coming. Quickly, what do you watch?

Roger: Activist investors or private equity circling a niche leader. A vendor whose growth story has stalled but whose installed base is huge and captive. Product investment visibly slowing while margins get managed up. Channel programs shrinking. When two or three of those line up on a vendor you depend on, that is the moment to fix your contract terms and warm your exits, not the day the press release lands. The press release is the notice window of acquisitions, and by now everyone at this table knows what happens when you wake up inside the window.

Bella: Round to close. One sentence for the VMware customer, or the customer of the next VMware. Bill.

Bill: Score your embedded vendors for the acquisition profile and negotiate change of control terms now, while they are cheap, because the day the deal closes, the catalog is the negotiation.

Laura: Run the partial exit. A slice of your estate on an alternative converts your walkaway from a brochure into a demonstrated capability, and it is available at every company size.

Roger: If you hold perpetual licenses, remember you own them. Third party support is a bridge that buys years of calm, and the option the vendor hates most is usually the one worth pricing first.

Bella: And mine. The catalog change lands heaviest on whoever can defend themselves least, so the under resourced mid market needs this discipline most and has it least. Next week, the seat that isn't a seat: consumption pricing, AI agents, and what happens to negotiation when the thing you are buying stops being a person. This is Off List. Read the paper before you sign it.