The key question
Does the VMware subscription conversion hold — or does shelfware and enterprise pushback erode the .5B EBITDA target?
VMware is the textbook Hock Tan acquisition: buy an undermonetized franchise, convert to subscription, cut costs ruthlessly, and raise prices. Operating margins leapt from 13-22% under VMware standalone to 77% under Broadcom, and the EBITDA target of $8.5B was achieved within three years of closing. The subscription transition is largely complete with 90% of top 10,000 customers signed onto VMware Cloud Foundation.
The tension in this segment is stark: Broadcom's financial execution has been extraordinary, but the customer backlash is real. A CloudBolt survey found 86% of enterprises actively reducing VMware footprint; AT&T filed suit over a 1,050% price increase; Gartner forecasts VMware share falling from 70% to 40% by 2029. The question is whether switching costs are high enough to keep churn manageable, or whether the price shock was so extreme that it accelerated migration timelines.
Switching costs vs. price shock: the core tension
VMware's embedded position in enterprise infrastructure creates enormous switching costs -- migrating workloads to Kubernetes or Nutanix is a multi-year project. But the 800-1500% price increases may have been aggressive enough to make that investment worthwhile for large enterprises. The 20% late renewal penalty suggests Broadcom knows the risk.
| VMware (AVGO) | Nutanix | Microsoft Hyper-V | Open Source (KVM/OVN) | |
|---|---|---|---|---|
| vSphere Market Share | ~70% enterprise hypervisor | ~15% | ~10% | <5% |
| NSX (networking) | Leading; 55K+ customers | Nutanix Flow (limited) | Azure-centric only | OpenStack (complex) |
| vSAN (storage) | Integrated; strong lock-in | Nutanix AOS (competitive) | Storage Spaces (limited) | Ceph (ops-heavy) |
| AI Integration | Broadcom Private AI; VCF AI stack | GPT-in-a-Box | Azure Arc / Copilot | None |
| Subscription Model | VCF subscription (forced migration) | All-subscription native | Per-seat M365 | Free; no vendor support |
| China Exposure | ~15-20% of VMware revenue (license) | Minimal | Restricted | Widely used |
| bull (25%) | base (50%) | bear (25%) | |
|---|---|---|---|
| EBITDA | Expands to $12B+ as AI features drive upsell | Sustains $8.5B+ with gradual NRR expansion | Compresses below $8.5B as churn exceeds renewal pricing power |
| Revenue | Grows 12%+ annually; add-on AI modules monetized | Grows 5-8% annually as VCF upsell scales | Flat to declining from FY2026 peak |
| VCF Retention | Enterprises convert shelfware; NRR expands to 115%+ | Most enterprises renew; shelfware converts over time | Shelfware churn materializes; China loss accelerates |
| DCF Value | $366B | $256B | $140B |
| Per Share | $77.3 | $54.1 | $29.6 |
| Bull Prob. | Bear Prob. | Implied Value | Δ from Current |
|---|---|---|---|
| 15% | 40% | $43.2/sh | -$11 |
| 20% | 35% | $46.7/sh | -$7 |
| 40% | 18% | $64.1/sh | +$10 |
| 35% | 22% | $60.4/sh | +$7 |
VMware is the floor of the AVGO thesis. At roughly $54/sh in the base case, it provides downside protection regardless of what happens to AI semiconductor revenue. The risk is not catastrophic: even in a severe churn scenario, VMware generates cash. The upside from AI-augmented VCF features is real but modest in the current model. The key watch item is the 2025-2027 VCF renewal cycle — when enterprise contracts signed at elevated pricing come up for renewal, the shelfware problem either resolves (enterprises find value and expand) or materializes (churn elevates, CEO admission proves prescient).
What is VMware's actual churn rate post-price increases?
The VMware integration exemplifies Hock Tan's acquisition playbook at its most effective: acquire an under-monetized franchise, convert to subscription, restructure the product portfolio into bundled offerings, raise prices aggressively, and cut costs. Operating margins expanded from the low teens to 77% in under two years, and the original $8.5B EBITDA target was achieved ahead of schedule.
Financial masterpiece, but growth is normalizing
The one-time benefits of subscription conversion and price increases are largely exhausted. FY2026 software guidance is low-to-mid single-digit growth. Future software value creation depends on VCF expansion, AI infrastructure positioning, and whether customer backlash translates into churn or just complaints.
The customer backlash against Broadcom's VMware pricing is among the most severe in enterprise software history. Survey data, lawsuits, and competitor win rates all confirm genuine customer outrage. However, the critical distinction is between intent and action: while 86% say they are reducing VMware footprint, actual migration is slow and painful. Deeply embedded VMware integrations take years to replace, and alternatives like Nutanix report migration challenges.
Intent vs action: the switching cost moat
The gap between customer anger and actual migration is Broadcom's moat. Enterprise VMware deployments are deeply integrated with networking, storage, and management workflows. Even with dramatic price increases, the total cost of migration (re-architecture, testing, downtime risk) often exceeds the price increase itself. This is why Broadcom can sustain 77% margins despite 86% of customers saying they want out.
Broadcom's infrastructure software segment is among the highest-margin software businesses in the world. At 93% gross margin and 77% operating margin, it converts nearly every dollar of revenue into profit. The $33.3B in remaining performance obligations provides multi-year revenue visibility, and the subscription model means revenue is almost entirely recurring. This segment is the financial foundation that allows Broadcom to invest aggressively in AI semiconductor growth.
The cash cow that funds the AI bet
Software generates nearly $21B in annual operating income at 77% margins, providing the cash flow to service acquisition debt, fund dividends, and invest in AI semiconductor R&D. Even if VMware churn increases, the subscription model and RPO provide years of revenue visibility. This is the most under-appreciated part of the Broadcom story.
Hock Tan's acquisition playbook is the most consistent and profitable in enterprise technology. Across three major acquisitions totaling over $116B, the pattern is identical: acquire under-monetized franchises with high switching costs, restructure to focus on large enterprise accounts, slash R&D and workforce, convert to subscription, and raise prices. The financial results are extraordinary -- every acquisition has seen dramatic margin expansion.
| CA Technologies (2018, $18.9B) | 300 layoffs immediately. Veracode sold to PE. R&D slashed. Mainframe portfolio optimized for margin extraction. |
| Symantec Enterprise (2019, $10.7B) | Cyber security services sold to Accenture within 6 months. Focus on enterprise accounts. Price increases on renewals. |
| VMware (2023, $86.3B) | EUC sold to KKR for $3.5B. OPM: 13-22% to 77%. Perpetual to subscription. 500-1050% price increases. 4 core bundles. |
Margin extraction vs product innovation
The fundamental question about Broadcom's playbook is sustainability. R&D cuts boost near-term margins but may erode long-term product competitiveness. If VMware Cloud Foundation fails to evolve while Nutanix and Kubernetes alternatives improve, the switching cost moat eventually erodes. VCF's positioning as an 'AI infrastructure abstraction layer' is the first real test of whether Broadcom can innovate alongside extracting.