What stays, what goes, what gets reissued in a Broadcom multi product exit plan.

Most buyers who arrive at the Desk with a question about exiting Broadcom are not asking whether to exit. They are asking how to reshape a contract footprint that has grown beyond the size the business can defend. The conversation is not migrate or stay. The conversation is which lines stay on the existing paper, which lines get partially reissued under new terms, which lines get fully reissued, and which lines get retired. The four buckets shape the rest of the negotiation. The buyer who walks into the renewal without a position on each of the four is negotiating against a seller who already knows the bucket the seller wants to keep the buyer in.

The framework that follows is the one the Desk uses to organise the decision. It is not a recommendation set. The buyer's situation determines which line goes in which bucket. The framework's value is that it forces the decision before the seller forces it the other way. A line that nobody on the buyer side has explicitly placed in a bucket will end up in the bucket the seller selects, which is almost always the most expensive bucket available.

The four buckets work across every product line the Desk covers. The detail varies. The shape does not.

Stay

A line stays on the existing paper when the steady state economics are acceptable, the entitlement footprint is reasonably aligned to deployment, the migration cost would be higher than the run rate savings, and the seller is willing to hold or improve the existing terms. The decision to stay does not mean the buyer renews without negotiation. It means the buyer renews on the same architecture, the same metric, and a renegotiated price.

The lines that most often stay across our file are the ones with low migration alternatives at scale. Mainframe software is the most consistent example. Brocade SAN is often a stay decision because the hardware refresh cycle dominates the economics. A subset of CA products that the buyer has heavily customised typically stays because the integration cost of moving exceeds the renewal cost of holding.

The negotiation for a stay line focuses on price, term, support tier and contract clauses. It is the most familiar of the four conversations. It is also the one where buyers most often concede on terms because the negotiation feels routine. The Desk's standing view is that even a stay line deserves the entitlement reconciliation and the calendar discipline that a full renegotiation requires.

Partial reissue

A line goes into partial reissue when the buyer wants to keep the product but on different commercial terms than the existing paper provides. The most common partial reissue is a metric change. The buyer moves from a per node basis to a per managed identity basis, or from a perpetual entitlement with maintenance to a fixed term subscription, or from a blended bundle to a line item itemisation that allows the buyer to retire components individually.

Partial reissue is the bucket where the Desk does the most precise work because the negotiation involves rewriting contract language. The Broadcom paper is templated and the seller's first answer to a partial reissue request is usually that the template does not accommodate the change. The buyer's leverage in this conversation is the alternative pathway and the credibility of the buyer's threat to escalate to a full reissue. A buyer who arrives with both produces a partial reissue that holds.

The lines that most often partial reissue across our file are Symantec endpoint moving from a perpetual maintenance basis to a subscription with a defined seat ceiling, VCF moving from a bundled subscription to an itemised subscription that allows the buyer to retire vSAN or Tanzu components individually, and CA workload automation moving from a blended capacity basis to a per workload metric that aligns to the buyer's actual usage.

"Partial reissue is the bucket where most of the buyer side value lives in 2026. The seller does not want the buyer to know it. The buyer has to ask for it explicitly. Nobody volunteers it."Contracts Lead, The Desk

Full reissue

A line goes into full reissue when the buyer wants to keep the product but on a wholesale new paper, usually under a different contracting vehicle, a different ordering pathway, or a different ownership structure inside the buyer organisation. The most common full reissue in our file is a buyer moving a piece of the footprint into a hyperscaler marketplace ordering pathway, which restructures the commercial relationship even though the underlying product remains the same.

Full reissue is operationally heavier than partial reissue but produces structural changes that partial reissue cannot. The buyer's relationship with the seller changes shape. The commercial vehicle changes. The renewal mechanics change. The audit clause typically changes because the new paper inherits the marketplace terms rather than the legacy Broadcom terms.

The lines that most often full reissue are VCF where the buyer is moving into a hyperscaler compatible footprint and the underlying VMware compute remains in place, Symantec endpoint where the buyer is restructuring the procurement vehicle through a security services partner, and Carbon Black where a similar pathway exists. The full reissue does not always reduce cost. It almost always changes the buyer's optionality on the next renewal.

Retire

A line goes into retire when the steady state economics no longer justify the run rate, the deployment has shrunk below a defensible threshold, or the line is duplicated by a capability the buyer already has elsewhere in the estate. Retire is the bucket buyers most often skip because the decision feels final. The decision does not have to be final. A line that retires from the Broadcom paper can be replaced with an alternative, repatriated to an internal capability, or simply switched off if the business need has dissolved.

The lines that most often retire across our file are perpetual CA entitlements that the buyer has been maintaining but not deploying, Symantec DLP perimeter components that have been duplicated by a Microsoft Purview rollout, and Brocade hardware support on legacy fabric that has been decommissioned but never removed from the support contract. The savings on retire decisions are often the cleanest in the negotiation because they require no replacement spend.

How to assign lines to buckets

The Desk uses a simple two by two to start the assignment. On one axis, is the steady state economics acceptable. On the other, is the contract architecture acceptable. A line where both are yes is a stay candidate. A line where the economics are acceptable but the architecture is not is a partial reissue candidate. A line where the architecture is acceptable but the economics are not is a full reissue candidate. A line where neither is acceptable is a retire candidate.

The two by two is not the answer. The two by two is the starting position for the conversation. Many lines move between buckets as the assessment progresses. The discipline of the assignment is the work. The buyer who completes the assignment for every line on the paper before the first commercial meeting walks into the negotiation with a position on every line. The seller cannot move the conversation onto undefended lines because there are no undefended lines.

The numbers in summary

What we have seen on live deals

The buyers who produce the best outcomes are the ones who treat the assignment as a discipline rather than an exercise. The four bucket framework is not novel. The novelty is the rigour of completing it before the first commercial meeting. Buyers who arrive with a position on every line negotiate from a different posture than buyers who arrive with positions on the top five lines and a vague intent on the rest. The seller can read the difference inside the first twenty minutes.

The second pattern is that the retire bucket is the most underused bucket in our file. Buyers tend to default lines into stay when the analysis would support retire, because the decision to retire feels operationally heavier than it actually is. The Desk's standing recommendation is to test every line for retire first, then partial reissue, then full reissue, and to default to stay only when the other three have been actively ruled out.

The third pattern is that the seller's posture on a full reissue is often softer than the buyer expects. The Broadcom desk is generally willing to accommodate a marketplace pathway or a partner pathway when the alternative is a full migration off the product. The buyer who asks the question gets an answer the buyer can use. The buyer who does not ask the question gets a renewal on the existing paper.

The Desk's exit planning work walks through the four bucket assignment line by line. The same discipline shows up in the Symantec practice and across every other Broadcom product line the buyer holds.

The takeaway

• Force every line into one of four buckets before the first commercial meeting. Stay, partial reissue, full reissue, retire. Undefended lines end up in the seller's preferred bucket.
• Test for retire first. Buyers default lines into stay when the analysis would support retire, and the savings on retire decisions are the cleanest in the negotiation.
• The partial reissue conversation is where most of the 2026 buyer side value lives. Ask for the metric change, the itemisation, the term shift. The seller will not volunteer it.