Diversifying after Broadcom.

Vendor diversification is the phrase enterprises reach for when a single supplier becomes too central to the cost structure, too dictatorial in the commercial relationship, or too important to operational continuity. In the case of Broadcom and VMware, the diversification conversation has moved from boardroom abstraction to active operational planning across many of the customer engagements we observe. The question is no longer whether to diversify; the question is what diversification actually looks like, where it pays back, and where it does not.

This article is about the practical mechanics of vendor diversification after Broadcom. It walks through the patterns we see working, the patterns we see failing, and the framework for deciding which path fits a given enterprise. It is opinionated about the trap of diversification as a posture rather than diversification as an outcome.

Why diversification is on the agenda

Several factors have moved Broadcom diversification from option to obligation for many enterprises. The headline subscription pricing increase materially shifts the long-term economics of staying on VMware. The bundle structure forces customers to license components they may not need. The audit posture creates compliance exposure that compounds over time. The product roadmap has narrowed in ways that make VMware less attractive as the exclusive infrastructure platform.

None of these factors individually require diversification. Together, they produce a strategic question that enterprises responsible to boards and CFOs cannot leave unanswered. The customers who arrive at a deliberate answer — diversify on these specific dimensions, stay focused on Broadcom on these others — produce better outcomes than the customers who default to either pure consolidation or unfocused diversification.

What diversification actually means

Vendor diversification in the Broadcom context is not a single strategy. It is at least four distinct strategies, each with different cost profiles, execution risks, and commercial benefits.

Hypervisor diversification

The most direct form of diversification is moving workloads from VMware vSphere to an alternative hypervisor — typically Hyper-V, Proxmox VE, Nutanix AHV, or KVM-based platforms. Hypervisor diversification reduces the VMware licence count, which directly reduces the Broadcom subscription cost.

The execution cost is non-trivial. Application validation, networking and storage adaptation, operations tooling, and skills development all add up. The customers who do this well typically migrate specific workload categories (test/dev, scale-out web tiers, batch workloads) rather than attempting wholesale hypervisor replacement. The customers who struggle are the ones who scope hypervisor diversification too broadly and find the execution cost exceeds the licence saving.

Public cloud diversification

A second form is repatriating selected VMware workloads to native public cloud services — Azure VMs, AWS EC2, Google Compute Engine — rather than keeping them on VMware regardless of where the VMware runs. This is different from running VMware on AWS or Azure as a managed service; it is moving the workload off VMware altogether.

The execution cost varies widely by application. Cloud-friendly workloads (stateless services, modern data platforms, containerised applications) migrate at reasonable cost. Legacy workloads with deep VMware-specific dependencies migrate at higher cost. The financial benefit depends on the comparison between VMware on-premises run cost and native cloud run cost at the customer's scale.

Adjacent stack diversification

The third form is diversifying the adjacent stack — storage, networking, monitoring, automation — to avoid the VCF bundle pull-through. Customers who use NetApp or Pure storage instead of vSAN, Cisco or Arista networking instead of NSX, Datadog or Dynatrace monitoring instead of Aria, are diversifying the stack without diversifying the hypervisor.

This form is often underestimated. The Broadcom commercial pressure is partly about the bundle effect; customers who hold credible alternatives across the bundle components have stronger commercial leverage at renewal even if they stay on vSphere. The diversification of adjacent components keeps the customer commercially flexible in ways pure hypervisor consolidation does not.

Operational diversification

The fourth form is diversifying the operations capability — keeping VMware skills strong while building skills in alternative platforms in parallel. This is the slowest and most expensive form of diversification, but it is the foundation that makes the others possible at scale.

Organisations that have only VMware-skilled operations teams cannot credibly migrate to alternatives even when the financial case supports the migration. Organisations that have built operational depth in alternatives can execute the other diversification strategies when the commercial conversation requires it.

The diversification trap

The most common mistake we see is treating diversification as a posture rather than as an outcome. The customer announces a multi-vendor strategy, communicates it to Broadcom, and then fails to actually execute meaningful workload movement. Broadcom sees through this immediately, and the commercial leverage the customer hoped to generate disappears.

Credible diversification requires actual executed movement. The customer who has actually migrated 500 VMs off vSphere is in a different commercial position than the customer who has talked about it for two years. The credibility of the alternative is the source of the commercial leverage; the announcement is not.

This produces a sequencing question. Customers who start with a small but real migration — even at a slightly negative ROI — to establish the operational capability and the credible alternative, are typically positioned better in the broader Broadcom relationship than customers who try to optimise the financial case for migration in isolation. The credibility premium is real and durable.

Sequencing the diversification

The sequence of diversification matters as much as the strategy. Customers who try to do everything simultaneously typically achieve none of it well; customers who sequence the work make material progress on each dimension.

A workable sequence we have seen is to start with adjacent stack diversification (lower execution cost, faster credibility), follow with operational diversification (slower, but enables the next phase), then proceed with selected hypervisor diversification on workloads where the financial case is positive, and consider public cloud diversification last for the workloads where it makes economic sense.

The sequence is not rigid; specific customers will diverge from it based on workload mix, existing skills, and commercial timeline. But the principle of staged, executable diversification rather than simultaneous large bets is durable across the customers we have observed.

The commercial conversation with Broadcom

Diversification produces commercial leverage only if Broadcom believes the customer is genuinely diversifying. The conversation with Broadcom about diversification needs to be evidence-based: this many VMs migrated to date, this much adjacent stack rationalised, this much investment committed to operational depth in alternatives.

Customers who walk into the Broadcom conversation with this evidence get materially different terms than customers who arrive with a strategy document. The evidence is the source of leverage; the strategy is the framing.

The conversation also needs honest framing. Broadcom is aware that most enterprises will not migrate the bulk of their VMware estate in the next three years. The credible position is not "we are leaving VMware"; it is "we are reducing our exposure and increasing our optionality, and we expect terms that reflect a customer with credible alternatives". This framing is defensible and produces the commercial response that pure rhetoric does not.

The cost of not diversifying

Customers who decide explicitly not to diversify also need a defensible position. The cost of staying single-vendor on Broadcom over a five-year horizon is large enough that the decision should be deliberate, costed, and reviewed annually. Customers who default into staying single-vendor without making the explicit decision routinely pay more than customers who make the decision explicitly with eyes open.

The defensible single-vendor position is to negotiate the Broadcom relationship aggressively, invest in the relationship management capability that produces good terms, and accept the cost premium as the price of operational continuity and consolidation benefit. This is a reasonable strategy for many enterprises; it is the strategy of last resort for enterprises that did not consider the alternatives.

Common diversification patterns

Across customer engagements, several diversification patterns recur frequently enough to be worth noting.

Pattern 1: Test/dev hypervisor migration. Moving test and development workloads to a lower-cost hypervisor reduces Broadcom exposure on the workloads that have the least operational risk to migrate. This is a typical first move for customers diversifying for cost reasons.

Pattern 2: Hyperscaler-native migration of cloud-friendly workloads. Workloads that already use cloud-native patterns (modern data platforms, containerised services) often migrate to native cloud services at reasonable cost, reducing VMware exposure on the most migration-ready workloads.

Pattern 3: Adjacent stack rationalisation. Replacing VCF bundle components with best-of-breed alternatives reduces the bundle pull-through cost. This is often the highest-ROI diversification because the execution cost is bounded and the licence saving is direct.

Pattern 4: Multi-cloud-via-VMware reduction. Customers who run VMware on AWS, Azure, or GCP for workloads that could run natively often find they are paying for VMware twice — on the hyperscaler and indirectly through the on-premises commitment. Rationalising this is a common diversification pattern.

What works and what does not

Vendor diversification works when it is executed deliberately, sequenced sensibly, and integrated with the broader commercial conversation with Broadcom. It does not work when it is announced as a posture without execution, scoped too broadly to execute, or treated as a one-time decision rather than an ongoing capability.

The enterprises we see handling this well treat diversification as a multi-year capability investment, not a project. They build the operational skills, establish the executed track record, maintain the commercial framing, and revisit the strategy annually. The compounding effect over three to five years produces a materially different commercial position from where they started.

The enterprises we see struggling treat diversification as a single decision and then under-execute it. The result is a strategy on paper and a status quo in practice, which produces the worst of both worlds: the cost of distraction without the benefit of credible alternatives.

The bottom line

Vendor diversification after Broadcom is real strategic work, not a slogan. The framework that works is deliberate, sequenced, executed, and integrated with the commercial relationship. The framework that does not work is rhetorical, unfocused, and disconnected from the actual Broadcom negotiation.

For most enterprises with material VMware estates, some form of diversification is the right answer over a five-year horizon. The specific form depends on the workload mix, the existing skills, the financial position, and the commercial timeline. The decision deserves the same rigour as any other major architectural commitment, and the payoff — measured in commercial leverage, financial flexibility, and operational resilience — is durable in ways the headline numbers around Broadcom do not capture.