Two years on from the Broadcom acquisition, the second and third VMware renewals are landing — and the sticker shock has not worn off. Perpetual licences are gone, the catalogue has collapsed into a few bundles, pricing is per-core with a minimum, and late renewals carry a penalty. The reflex is to either grit your teeth and pay or rip everything out. Both are usually wrong. This is the decision framework we walk UK clients through: what changed, how to model the three realistic paths, and how to use a credible exit to cut the number even if you choose to stay.
What actually changed
Broadcom moved every customer to subscription and folded the wide product range into a small number of bundles led by VMware Cloud Foundation. Pricing shifted from per-CPU-socket to per-core, with a per-host core minimum, and late renewals attract a penalty. For some organisations the headline renewal number has multiplied several times over. The licensing model now rewards dense, well-utilised hosts and punishes sprawl — which is the first lever, before you even discuss alternatives.
The practical effect is that your VMware cost is now tightly coupled to core count and bundle tier. Consolidating onto fewer, denser hosts and dropping to the bundle you actually use can take a meaningful bite out of the quote on its own.
Path 1 — negotiate (and mean it)
Broadcom discounts against leverage, and leverage is a credible alternative. Buyers who model moving even a portion of the estate to an alternative — and can show the migration is real — have recovered double-digit percentages off the opening quote. The work is in the modelling: know your core counts, your true bundle needs, and the cost and effort of an exit, so the conversation is evidence-based rather than a bluff.
Negotiation is the right primary path when VMware is deeply embedded (NSX, vSAN, Aria, strict ISV support matrices) and the switching cost genuinely exceeds the saving. But you only get the discount if the exit is believable.
Path 2 — re-platform
For many mid-market estates the alternatives are now mature enough to be the better answer, not just the threat. The front-runners we see in UK deployments:
- •Nutanix AHV — the most enterprise-proven full-stack replacement, strongest where you're replacing VMware Cloud Foundation and want a one-click-style migration.
- •Proxmox VE — open-source (KVM-based) with HA clustering, live migration and backup built in, no licence cost; strong for cost-led, technically-confident teams.
- •Microsoft (Hyper-V / Azure Local) — a natural fit where you're already Microsoft-centric and want one vendor relationship.
- •VMware retained (right-bundled) — sometimes correct if the workloads and ISV support genuinely require it; the trick is buying only the bundle you use.
Path 3 — exit to cloud or hybrid
A renewal is also a prompt to ask whether some workloads belong on-premises at all — and whether others would be cheaper repatriated from cloud. This is rarely all-or-nothing: the strongest outcomes are hybrid, moving the workloads that suit each model and using the renewal as the forcing function to finally do the analysis.
Whichever path you choose, start 9-12 months before the renewal date. The organisations that negotiate from weakness are the ones who left it too late to build a credible alternative; time is the real leverage.