What Broadcom migration economics actually look like across the portfolio in 2026.

Every buyer the Desk works with this year is asking some version of the same question. What does it actually cost to leave a Broadcom product line in 2026, and what does it cost to stay. The answer matters even for buyers who have no intention of leaving, because the number that comes back determines the buyer's posture at the renewal table. A buyer who can quote a defensible exit cost on the back of an envelope negotiates differently than a buyer who cannot. The Broadcom desk knows the difference and prices accordingly.

This piece is the Desk's working view of what migration economics actually look like across the six product lines we cover. The numbers are drawn from costed alternative pathway assessments we have commissioned on behalf of buyers in the last twelve months, paired with reported steady state economics from buyers who completed migrations and from buyers who started them and stopped. The piece is not a recommendation to migrate. It is a recommendation to know what migration would cost before signing the next renewal.

The pattern that holds across all six lines is that the steady state economics of migration are almost always more favourable than the buyer's intuitive estimate, and the transition cost is almost always less favourable. The buyer who only models the steady state will reach for migration too readily. The buyer who only models the transition will reach for renewal too readily. The buyer who models both arrives at the right conversation at the right time.

VCF and the VMware estate

VCF migration economics are the most studied case in our file because they are the most contested in active engagements. The two credible alternatives at scale are Nutanix and a managed hyperscaler footprint with a VMware compatible compute layer. A third option, Red Hat OpenShift with KubeVirt, has matured enough to enter serious consideration in the last twelve months but remains less common as a wholesale replacement.

The transition cost of moving a VCF estate of two to four thousand VMs to Nutanix lands in the range of three to five times the annual VCF run rate. The steady state cost after migration lands in the range of 55 to 70 percent of the post acquisition VCF run rate. Payback against the migration cost happens between year three and year five depending on how the destination is sized. The window matters because it falls inside the typical refresh cycle of underlying hardware, which means the hardware cost is partially absorbed by a refresh that would have happened anyway.

The hyperscaler pathway has a smaller transition cost on paper but a larger long run cost because compute rates at scale tend not to favour a like for like lift. The pathway works best when the buyer is moving a subset of the estate rather than the whole estate, and when the subset has elastic demand characteristics that the hyperscaler pricing rewards.

Symantec endpoint and DLP

Symantec endpoint migration to CrowdStrike or SentinelOne is the most operationally mature pathway in the file. The transition cost lands in the range of one to two times the annual Symantec run rate, mostly in deployment services, identity reconciliation and policy translation. The steady state cost varies more than the VCF case because the alternatives price aggressively for displacement and the buyer can often anchor the negotiation on the prior Symantec spend. The steady state cost lands anywhere from 60 to 95 percent of the post acquisition Symantec run rate.

DLP migration is meaningfully harder than endpoint migration. The destination market is thinner. Microsoft Purview is the most common alternative we see considered but the policy translation work and the integration with the buyer's existing identity perimeter is not trivial. The transition cost lands in the range of two to four times the annual DLP run rate. The steady state cost lands in the range of 50 to 75 percent. The Desk's standing view is that DLP migration is rarely cost positive inside a three year window but the costed assessment still has value at the renewal table.

Carbon Black

Carbon Black EDR migration follows a similar shape to Symantec endpoint with slightly different numbers. The destination set is the same. The transition cost is somewhat lower because the Carbon Black deployment patterns are typically smaller and more contained. The steady state cost lands in a similar range to Symantec endpoint. The complication is the Carbon Black Cloud Workload and container coverage, which has fewer credible alternatives at the same maturity. Buyers with a heavy container footprint find the assessment more nuanced. Buyers with a primarily endpoint footprint find it straightforward.

"The migration model is not the negotiation strategy. The migration model is what makes the negotiation possible. Without it, the buyer cannot answer the seller's most important question."Exit Planning Lead, The Desk

CA Technologies

CA migration economics are the most varied in the file because the CA portfolio inside Broadcom contains products with very different replacement profiles. CA AIOps has credible alternatives in the modern observability set. CA workload automation has alternatives that range from competing schedulers to a partial replacement with cloud native orchestration. CA identity has a longer tail of replacement options at different price points. The transition cost across the CA set typically lands in the range of two to four times annual run rate but the variance is wide.

The steady state economics are usually favourable on a per capability basis but the migration tends to require the buyer to accept some functional loss or to fund integration work that did not exist under the consolidated CA contract. Buyers who plan the migration around a specific product line within CA reach a cost positive outcome more reliably than buyers who attempt a wholesale CA exit.

Mainframe software

Mainframe software is the line where the migration question is most often answered by saying that migration is not the right question. The credible alternatives for the mainframe software estate, which spans CA mainframe and a number of products acquired through the original acquisition, are not displacement options. They are gradual workload reshape options that move discrete components off the mainframe over a decade rather than years. The transition cost is functionally indefinite. The renewal posture for mainframe software depends less on the alternative pathway and more on the capacity squeeze and the support basis. The Desk treats mainframe renewals as a different shape of engagement for this reason.

Brocade SAN

Brocade SAN migration economics are dominated by the end of life conversation on legacy fibre channel hardware. The credible alternatives are a hardware refresh inside the Brocade portfolio, a migration to a competing fabric vendor, or a workload reshape that reduces SAN dependence. Each pathway has a different cost profile and the choice depends heavily on the buyer's storage refresh cycle. The Desk's standing view is that the Brocade conversation is best decoupled from the broader Broadcom portfolio negotiation because the underlying technology decisions are independent.

The numbers in summary

What we have seen on live deals

The most consistent pattern in our file is that the buyer who arrives at the renewal table with a costed pathway in writing produces a better outcome than the buyer who arrives with no pathway at all, even when the buyer has no intention of using the pathway. The concession band on the renewal opens by 18 to 34 percent when the pathway is visible. The cost of the assessment is recovered many times over in the renewal even when no migration is triggered.

The second consistent pattern is that buyers who attempt to model migration economics from public benchmarks alone arrive at numbers that are unreliable in either direction. Vendor published benchmarks tend to overstate the favourable case. Anecdotal cost stories tend to overstate the transition burden. The cost of a proper assessment, in the range of forty thousand to one hundred and twenty thousand dollars depending on scope, produces numbers that survive contact with both the buyer's finance team and the seller's commercial team.

The third consistent pattern is that the timing matters. Assessments commissioned six months before renewal produce different conversations than assessments commissioned thirty days before renewal. The earlier work feeds into the buyer's calendar, the buyer's posture and the seller's perception of the buyer's optionality. The later work feeds into nothing except the negotiation itself, which is too late to shift the seller's opening position.

The Desk's exit planning work is built around these patterns. The output is not a migration plan. The output is the document the buyer carries into the renewal. The shape of that document is the same regardless of which line in the VMware practice or any other Broadcom product line the buyer is negotiating.

The takeaway

• Commission the costed pathway assessment six months before renewal. The assessment changes the seller's opening posture, not because the buyer intends to migrate, but because the seller knows the buyer could.
• Model both the transition cost and the steady state cost. Buyers who only model one of the two reach for the wrong decision in either direction.
• Treat mainframe and Brocade as separate conversations. The economics of those two lines do not behave like the rest of the Broadcom portfolio and the negotiation posture is different.