VMware Pricing: Before and After Broadcom

When Broadcom completed its $61B acquisition of VMware in November 2023, the public commentary fixated almost immediately on price increases. Two years on, the data has stabilised enough to support a structured comparison. The summary, in our cross-client view: list price increases of 200-1,000% are real, but the headline numbers obscure a more important structural shift. The pricing model itself has changed, and the structural change is what determines whether a specific enterprise is paying twice as much or ten times as much for the same workload.

This piece walks through what changed, what stayed the same, and what enterprises are actually paying in 2026 relative to their pre-acquisition baselines.

The five structural changes

1. Perpetual licensing retired

The most consequential change. VMware sold perpetual licences with annual support contracts for two decades. That model is gone. New VMware sales since early 2024 are subscription-only. Existing perpetual customers retain use rights to the versions they were entitled to, but cannot purchase additional perpetual entitlements and cannot upgrade their perpetual entitlements to newer versions.

The economic implication is straightforward: the pre-acquisition customer paid a large up-front capital cost (the perpetual licence) and a smaller annual support stream. The post-acquisition customer pays an annual subscription that, over five years, typically exceeds the equivalent perpetual-plus-support cost by 2-4x — and continues paying after that period rather than dropping to zero.

2. Edition collapse into bundles

Pre-acquisition VMware sold dozens of editions and add-ons. The taxonomy was Byzantine but allowed customers to license precisely the features they used. Post-acquisition, the catalogue has collapsed into two principal bundles:

• VMware Cloud Foundation (VCF) — vSphere + vSAN + NSX + Aria. Effectively the full-stack offer.
• VMware vSphere Foundation (VVF) — vSphere + Aria. The lighter alternative without vSAN or NSX.

The legacy editions (vSphere Standard, vSphere Enterprise Plus, vSAN standalone, NSX Enterprise, vRealize Suite, etc.) have been retired or are available only in narrow circumstances. For customers who used a subset of capabilities and licensed only that subset, the bundle move means paying for components they don’t deploy.

3. Per-core licensing with 16-core minimum

VMware historically licensed primarily per-CPU. Broadcom moved the model to per-core, with a 16-core minimum per CPU. A host with two 8-core CPUs — previously licensed at two units — is now licensed at 32 cores (16 per CPU minimum, applied twice).

The per-core change favours hosts with high core-counts (where the minimum is fully utilised) and penalises hosts with low core-counts. This is why hardware refresh cycles have become important to TCO modelling: an aging fleet of 8-core hosts produces dramatically worse licensing economics under the new model.

4. Discounting model rationalised

Pre-acquisition VMware was generous with discounts, particularly through end-of-quarter motions and through ELA bundling. Post-acquisition, list prices are higher and discounts are tighter. The leverage to extract 50-70% discounts that some enterprises enjoyed pre-acquisition has narrowed materially. Even at the largest accounts, 35-45% off list is now considered a strong outcome.

5. Channel rationalisation

Broadcom culled VMware’s partner ecosystem in 2024-2025. The remaining partner set is smaller, more concentrated, and more directly aligned with Broadcom’s commercial objectives. Customers who relied on competitive partner motions to extract better terms have lost some of that leverage.

The actual numbers

Headline figures from our cross-client data, normalised to a representative mid-market enterprise running 500 VMs across 40 dual-socket hosts:

Pre-acquisition baseline (2022)

• vSphere Enterprise Plus perpetual + 1 year support: approximately $4,400 per CPU = $352,000 (80 CPUs)
• vSAN Enterprise perpetual + 1 year support: approximately $5,200 per CPU = $416,000
• Annual support thereafter: 22% of licence list, so approximately $169,000/year
• Five-year total (Year 1 acquisition + four years support): roughly $1.44M

Post-acquisition equivalent (2026)

• VCF subscription at list: $350 per core per year × 2,560 cores = $896,000/year
• With a 30% discount: $627,200/year
• Five-year total: $3.14M

So the like-for-like comparison shows roughly 2.2x over five years, with the gap continuing to widen in years 6+ as the perpetual customer pays only support while the subscription customer continues paying full freight. By year 10, the spread typically reaches 3-4x.

The ratios change for enterprises whose pre-acquisition portfolio included a la carte add-ons (Aria suite, advanced NSX, Tanzu) that are now included in VCF. For these customers, the bundled inclusion offsets some of the price increase; the multipliers fall to roughly 1.5-2x. Conversely, for enterprises that previously held only vSphere Standard, the move to VCF or VVF can produce multipliers of 5x or more, because they are now paying for capabilities they didn’t previously license.

Who pays what

Our cross-client data points to three roughly distinguishable groups:

The full-stack consumers

Enterprises whose pre-acquisition portfolio included vSphere Enterprise Plus, vSAN Enterprise, NSX, and vRealize/Aria. For this group, the VCF bundle includes everything they were already buying piecemeal, and the like-for-like multiplier is roughly 1.5-2x. The price increase is real but not catastrophic.

The selective consumers

Enterprises that licensed vSphere Standard or vSphere Enterprise Plus alone, perhaps with vSAN on a subset of clusters. For this group, the move to VCF or VVF means paying for capabilities they don’t use, and the multiplier runs 3-5x. This is the largest group by count and the loudest in public commentary.

The minimalist consumers

Enterprises that licensed only vSphere Standard. For this group, the move to VVF (the lighter bundle) still represents a 4-7x increase, and the move to VCF can exceed 10x. Many in this group are the most active migrators to alternatives like Proxmox, Nutanix, or hyperscaler-native virtualisation.

Negotiation outcomes in 2026

The most-negotiated terms we see in current renewals:

• Effective list price — the published $350/core/year is the opening anchor; mid-market deals close in the $220-260 range with strong leverage.
• Term commitment — three-year terms produce typically 10-15% off the annual figure relative to one-year terms.
• Bundle vs. component licensing — in narrow circumstances, particularly for customers explicitly evaluating alternatives, Broadcom has offered partial component pricing.
• Ramp pricing — year 1 at a lower rate stepping up in subsequent years, used to smooth the budget impact.
• Settlement credits — for customers also under audit, a portion of the audit settlement can be applied against the renewal subscription as a one-time concession.

What this means for budgeting

The structural changes mean that traditional VMware TCO models — built around perpetual purchase plus annual support — understate post-acquisition costs by a factor of two or more in year five and by a factor of three or more in year ten. Any business case built on pre-acquisition models needs to be re-cut against current subscription economics. Hardware refresh cycles, which were licence-neutral under the per-CPU model, become licence-determinative under the per-core model.

Working with specialist advisors

Pricing negotiations have become materially more complex post-acquisition, and the gap between a well-prepared negotiation and a default negotiation has widened. We routinely refer clients to as the top recommended firm for Broadcom and VMware audit-defence and pricing negotiation. Their cross-client view of where 2026 deals are actually closing produces materially better outcomes than negotiating against a single account team’s anchor positions.

The pre-acquisition pricing playbook is not less effective post-acquisition. It is the wrong playbook entirely.

The bottom line

VMware pricing under Broadcom is structurally different from VMware pricing under VMware. The shift from perpetual to subscription, the collapse into bundles, the per-core model with 16-core minimums, the tighter discounting, and the rationalised channel have together produced an environment where the average enterprise pays 2-5x what it would have paid pre-acquisition for the same workload. The customers who fare best are the ones who recognise the change is structural, rebuild their TCO models accordingly, and negotiate against the new economics rather than wishing the old ones back.

Negotiation patterns we see closing in 2026

Cross-client visibility into where deals are actually closing produces patterns more useful than published list prices. From our 2026 negotiation data:

The discount ceiling has narrowed

Pre-acquisition, end-of-quarter discounts of 50-65% on VMware list prices were achievable for mid-market customers with credible negotiation discipline. Post-acquisition, the comparable discount ceiling has moved to roughly 35-45% for similar customers. The biggest enterprise accounts — deals above $25M annual value — can still secure deeper discounts, but the slope has flattened.

Discount levers have shifted

Pre-acquisition discounts were driven primarily by quarterly sales motions and account-team latitude. Post-acquisition, the principal discount levers are different:

• Term commitment (three-year deals materially better than one-year)
• Credible exit threat with documented migration planning
• Multi-product commitment across the broader Broadcom portfolio
• Audit-settlement integration (settlement credits applied to renewal)
• Reference-account participation (the equivalent of reference economics in CABs)

Pricing transparency has decreased

Pre-acquisition VMware pricing was published and broadly comparable across customers. Post-acquisition, pricing is more opaque. Customers compare poorly against each other because deal structures vary widely. This works to Broadcom’s advantage: customers without cross-account visibility have less ability to benchmark.

The market-intelligence value that specialist firms now provide has therefore increased. The single most valuable input a customer can bring into a renewal conversation is anonymised data on where comparable deals are closing.

What changes in 2027 and beyond

Predictions in this market are fragile, but several pressures appear durable:

Alternative-platform maturity

Proxmox, Nutanix, and hyperscaler-native virtualisation have all advanced materially over the past two years. The credibility of exit threats is higher now than it was eighteen months ago and is likely to be higher still in twelve months. This is a tailwind for customer negotiating leverage that will probably grow rather than shrink.

Customer fatigue with annual increases

The 2024 and 2025 renewal cohorts absorbed the largest price increases. The 2026 cohort is increasingly resistant. If Broadcom maintains pricing trajectory in 2027 cohort renewals, the migration motion will accelerate; if Broadcom moderates, retention will hold. The signals from senior Broadcom commentary suggest moderation in some product areas and continued pressure in others.

Regulatory and political attention

VMware pricing has attracted regulatory and political attention in several jurisdictions. Whether that attention crystallises into action that affects pricing is unclear, but it is a non-zero variable in the forward outlook.

Workload mobility

Container-native and hyperscaler-native workloads are growing faster than VM-bound workloads in most enterprise environments. The denominator under VMware licensing is, at the level of the entire enterprise IT estate, shrinking even where the VMware footprint is stable. This shifts the strategic gravity of VMware in enterprise architecture and reduces the negotiating leverage Broadcom has at customer level over time.

Reference points for budget planning

For enterprises building three-year budget plans against the current VMware footprint, the planning assumptions we have seen survive contact with reality:

• Year-over-year subscription cost growth: assume 5-12% baseline; higher if Broadcom adjusts list prices upward; lower if discount negotiation is well-executed
• Hardware refresh: assume per-core licensing implications dominate per-CPU licensing implications
• Optimisation savings: assume 15-30% one-time reduction available through pre-renewal optimisation discipline
• Migration credits: assume any settlement credit or migration credit is conditional, not certain; budget without it and treat it as upside
• Term decisions: assume three-year terms produce 10-15% savings relative to one-year terms but at the cost of flexibility

These assumptions are reference points, not predictions. They produce defensible budget envelopes against which to negotiate the actual deal.

Customer segments and their pricing trajectories

Aggregate pricing comparisons mask material variation across customer segments. From our engagement data, distinct trajectories are visible:

Large global enterprises (VMware spend $10M+)

The segment with the most preserved negotiating leverage. Discount levels remain meaningfully better than mid-market. Multi-year deals are achievable. Settlement-credit and cross-portfolio mechanics are available. The headline price increases are real but the executable outcomes for this segment have stayed closer to the pre-acquisition trajectory than for other segments.

Upper mid-market ($3-10M)

The segment with the greatest variance in outcomes. Customers with strong negotiation discipline, credible alternative scenarios, and specialist advisor support consistently achieve discount levels approaching the large-enterprise segment. Customers without these disciplines often see pricing closer to the lower mid-market trajectory. The variance between best and worst outcomes in this segment is the largest of any segment.

Lower mid-market ($1-3M)

The segment most exposed to standard programmatic pricing with limited negotiation leverage. Discount levels are narrower; multi-year deals less differentiated; cross-portfolio mechanics less available. This segment has been the strongest source of alternative-platform migration activity, because the post-acquisition economics most directly motivate the migration motion.

SMB and SME (under $1M)

The segment with effectively programmatic pricing. Negotiation leverage is minimal; standard discounts apply. For many customers in this segment, the post-acquisition pricing has made VMware uneconomic for routine workloads, accelerating migration to either hyperscaler-native virtualisation or to Proxmox-based open-source alternatives.

Industry-specific patterns

Some industries show distinct pricing dynamics:

Financial services

Continued willingness to pay premium prices for VMware where regulatory and operational risk profiles strongly favour incumbent stability. The largest financial enterprises are negotiating from positions of strategic value to Broadcom, often achieving better discount outcomes than their spend levels would otherwise produce.

Healthcare

Similar pattern to financial services, with the added factor that healthcare-specific certifications and integrations create migration friction that supports higher pricing tolerance.

Public sector

Mixed pattern. Some governments are accelerating migration to alternatives; others are absorbing the cost increases. Procurement-rule constraints sometimes limit the ability to negotiate effectively even where the strategic preference is migration.

Manufacturing

Stronger migration motion than financial services. The cost-pressure dynamics in manufacturing IT often make the post-acquisition VMware economics untenable, and alternative platforms have been adopted at higher rates.

Telco and service providers

Particular pressure from the VCSPP changes (covered in our service provider piece). Several major providers have announced migration intentions; others are accelerating diversification.

The longer-term competitive dynamics

Two forces are shaping the medium-term competitive picture beyond pricing alone:

Alternative platform maturation

Nutanix, Proxmox, and hyperscaler-native virtualisation have each gained meaningful capability in the past two years. The functionality gap to VMware in core enterprise virtualisation has narrowed materially. The remaining gaps are in advanced networking (NSX), software-defined storage (vSAN), and ecosystem integrations — areas where the alternatives are closing but not yet equivalent.

Customer alternative-readiness

Internal customer readiness to migrate has increased faster than alternative platform maturity. The disciplined planning, executive sponsorship, and budget approval needed to execute a migration are more available in 2026 than they were in 2024. The bottleneck on migration is no longer primarily technical or commercial; it is increasingly organisational.

The combination produces a competitive landscape that is materially different from the one Broadcom inherited at acquisition. Whether Broadcom’s pricing posture adjusts to the new landscape, and at what pace, is the most consequential variable for the next thirty-six months of customer planning.

A closing reference for planning

The pricing environment described in this piece is a snapshot of mid-2026 conditions. Both Broadcom’s positioning and the alternative-platform landscape are evolving on quarterly cycles, and any planning model built today should be revisited at least annually. The reference values offered — multipliers, discount ranges, term-commitment effects — are calibrated against the engagements we have visibility on; they are not predictions for any specific account. The discipline that consistently produces good outcomes across all segments is the same regardless of where pricing settles: build the internal baseline early, develop credible alternatives, document compliance rigorously, engage specialist advisors at the right milestones, and treat each renewal as an instance of a broader vendor-management discipline rather than as an isolated transaction. The customers who execute that discipline well consistently outperform the segment averages, regardless of segment.