TSA infrastructure decommissioning is the cost reduction discipline that turns a TSA cutover into actual savings. The cutover moves Newco off the seller's environment but the seller's infrastructure can keep running for months unless the decommission plan is explicit. Disciplined TSA cost reduction treats the decommission as a separate workstream with named owners, dated milestones, and a closeout report that proves the cost has actually dropped out of the run rate.
Cutover from a parent environment to Newco's standalone footprint typically leaves the parent's infrastructure intact. The seller does not turn off the systems on the cutover date because other business units may still depend on shared capacity. Storage stays mounted. Compute stays provisioned. Network paths stay open. The cost continues to flow through the TSA invoice as if Newco still consumed the capacity.
The disciplined buyer plans the decommission as part of the cutover plan, not as a separate phase. The cutover document includes a decommission section for each system. The section names the components to be retired, the owner for the retirement, the date for the retirement, the validation steps, and the cost reduction expected. Without this section, the cutover delivers operational independence but does not deliver the savings.
The seller's incentive structure matters. The seller is paid through the TSA for the capacity that remains active. The seller has no commercial incentive to accelerate the decommission. The disciplined buyer builds the decommission obligation into the TSA itself or into a side letter that locks the seller's commitment with named milestones. The pattern overlaps with the broader TSA stranded cost elimination approach.
The first task is the inventory. Every component that was provisioned for Newco use during the TSA period is listed. Compute instances, storage volumes, database instances, network circuits, load balancers, security appliances, backup repositories, monitoring agents, integration platforms, and management consoles. Each component has an asset tag, a cost basis, and a current consumer.
The inventory comes from three sources. The seller's CMDB if one exists. The TSA invoice with infrastructure line item detail. The cloud or virtualization platform's billing console for direct usage data. The three sources rarely match. The disciplined buyer reconciles them and uses the union of the three as the working list. Anything appearing in any of the three lists must be accounted for in the decommission plan.
Each entry gets one of four dispositions. Decommission outright, where the component is retired and the cost drops. Migrate to Newco's environment, where the workload moves but the seller's component is still retired. Reassign to another seller business unit, where the component continues but stops billing through the Newco TSA. Retain on the TSA, where the component continues to support Newco and the cost is justified. The deeper pattern lives in the TSA exit application cutover planning article.
Most cutovers include a parallel running window where the old and new environments operate simultaneously. The window protects against rollback. The window also costs money on both sides. The disciplined buyer caps the window in the cutover plan with a defined exit date and a named decision maker for any extension.
Parallel running of two ERP environments, two CRM environments, or two data warehouse environments can run several million dollars per month. The cost includes the seller's TSA charge for the legacy environment plus Newco's standalone cost for the new environment. The cost is acceptable for a defined window but indefensible as a standing posture. Every additional week of parallel running has to be justified against the rollback risk it actually mitigates.
The decommission of the legacy environment is then sequenced. Read only access first, write access cutover, full decommission. The full decommission requires the legacy environment to be backed up, archived, and retired. The seller's TSA charge for the legacy environment ends on the decommission date. The pattern overlaps with the broader TSA exit parallel running strategy approach.
Physical data center and colocation exits are the largest single decommission events. The carve-out may have inherited rack space in the parent's data center, dedicated cages in a colocation facility, or shared backbone capacity. The exit requires physical asset removal, secure data disposal, contract termination, and lease close down. Each step has lead time.
The colo contract is often the bottleneck. Colocation agreements have notice periods of 90 to 180 days and may include early termination fees. The disciplined buyer reads the contract, calculates the breakeven between continuing to pay and the early termination fee, and chooses the path that delivers the earliest savings. The fee is sometimes worth paying because the savings on rent, power, and bandwidth pay back the fee within three to six months.
Data sanitization at the exit is a security and regulatory matter. Storage media that held Newco data must be wiped or destroyed under documented procedures. The seller's TSA charge does not always include sanitization. The buyer should ask for the sanitization certificates as part of the decommission closeout. The pattern overlaps with the broader carve-out network separation work.
Cloud infrastructure decommissioning is different from physical decommissioning. The cloud bill drops the moment the resources are deprovisioned. The discipline is in finding the resources that should be deprovisioned. Test environments left running. Storage volumes attached to terminated instances. Snapshots that nobody owns. Reserved instance commitments that the parent locked in for capacity Newco no longer needs.
Reserved instance and savings plan commitments are a specific challenge. The parent's commitment was sized for the parent's usage including the carve-out. After the cutover, Newco moves to its own cloud account with its own commitments. The parent is left with reserved capacity that exceeds remaining demand. The TSA usually allows the parent to pass through the unused commitment cost. The disciplined buyer challenges the pass-through and negotiates a sharing arrangement at signing or in the renegotiation.
Right sizing of Newco's own cloud footprint is then a parallel exercise. The cloud bill should drop monthly during the first 90 days as orphaned resources are cleaned up and instance sizes are tuned to actual load. Specialist support on the entire decommission and right sizing program is part of the TSA Renegotiation service when the buyer needs the program managed at scale.
How buyers find and remove the costs that survive after the TSA ends.
Read the article →How buyers exit shared sites, leased space, and data center capacity tied to the parent.
Read the article →How buyers cut software license waste in the TSA without breaking critical access.
Read the article →The 90-day governance, IT, finance, HR and procurement separation plan we run on live carve-outs. Get the playbook plus the bi-weekly Day One Letter — short, signal-heavy, buyer-side.
No spam. Unsubscribe in one click. · Read the overview first →

Fixed-fee proposal in 48 hours. Senior team on day one. The first conversation is always free.
Seven buyer-side moves to exit a Transition Services Agreement on time and below budget. The mark-up, the extension-fee curve, exit sequencing, and the 11-month calendar.
One tactic, one benchmark, or one pattern from a recent buyer-side engagement. Short. Signal heavy. Free.
Subscribe to The Day One Letter →