Exiting the TSA cleanly on software.

Exiting the TSA cleanly on software is the buyer side work of leaving the parent transition services on a fixed date with every application re licensed, every user migrated, every integration rebuilt and every parent entitlement switched off, so the new entity stands entirely on its own with no stranded cost and no audit gap. The exit is the moment the carve out is finally complete, and it is also the moment most likely to go wrong. A TSA that drifts past its exit date keeps the new entity paying the parent and keeps it exposed to a publisher arguing that the licenses were never properly transferred. Planning the exit from day one is what turns a soft deadline into a clean break.

What exiting the TSA cleanly on software requires

A clean software exit is not a single event but the convergence of four conditions, all of which have to be true before the exit date. The new entity has to be licensed in its own name for every application it runs, so nothing depends on a parent agreement. Users and data have to be migrated onto the new entity systems, so access and records no longer live on the parent. Integrations have to be rebuilt so that feeds point to the new entity systems rather than the parent. And parent access has to be revoked, so there is no residual use of an entitlement that belongs to the seller. Miss any one of these and the exit is not clean, it is a handover with loose threads that turn into cost or an audit.

These four conditions are why the exit has to be designed at the start of the TSA, not improvised at the end. Re licensing needs publisher lead time, identity migration needs careful sequencing, and integration rebuilds need testing. A buyer that waits until the TSA is nearly over to start this work will overrun the exit date and pay for the privilege. The exit plan is the natural continuation of the TSA exit timeline and software critical path and the wider carve out software licensing playbook.

The milestones that prove the exit is done

A clean exit is provable, not assumed. Each of the four conditions resolves into a milestone with hard evidence behind it. Re licensing is complete when every application sits on a signed agreement in the new entity name, not when someone believes it does. Identity cutover is done when parent accounts are disabled and users authenticate only through the new tenant. Data migration is done when record counts reconcile on both sides. Integration rebuild is done when an end to end business process, traced from entry to outcome, runs entirely on the new entity systems. And access revocation is done when the parent access logs show zero use of the parent entitlements by the unit. The table sets out each milestone with the evidence that proves it, because an exit signed off on belief rather than evidence is the one that produces an audit finding later.

The access revocation milestone is the one buyers most often treat as a formality and most often get wrong. A service account that still authenticates against a parent system, a report that still pulls from the parent warehouse, or a user who still has a parent login is residual use of an entitlement the new entity no longer owns. Publishers treat that residual use as unlicensed deployment, and a carve out is exactly the kind of event that prompts them to count. Proving zero residual use, with access logs rather than assurances, is what closes the audit door on the way out.

Why a delayed exit costs more than the TSA fee

The obvious cost of a slow exit is the extending TSA fee, but that is the smallest part. The larger cost is the audit risk that grows with every month the new entity runs on parent licenses. Inherited and unresolved licensing of this kind is precisely what produces large post deal claims. SAP pursued Diageo for a reported 60 million over disputed and inherited licensing, as reported in trade coverage as of 2017, and the conditions that produced that exposure are the same conditions a drifting TSA creates. The second hidden cost is leverage. A new entity that has overrun its exit date and still depends on the parent has no negotiating position, with either the parent or the publishers, and pays accordingly.

The buyer answer is to drive the exit to a date and hold it. Every milestone should have an owner and a deadline that sits comfortably before the contractual exit, so there is room for the inevitable slippage without breaching the date. Where a milestone genuinely cannot be met, an extension should be negotiated deliberately and priced, not stumbled into. Treating the exit as a managed project rather than a hoped for outcome is the difference between a clean break and an entangled one, and it is the bridge to standing up the new entity software estate.

How an independent advisor drives a clean exit

An independent, buyer side advisor drives a clean exit because the advisor has no incentive to extend the TSA or to sell the new entity software it does not need. The advisor builds the exit plan from the dependency map, tracks every milestone to its evidence, runs the re licensing negotiations without the conflict a reseller carries, and confirms zero residual parent use before sign off. This is commercial and licensing work, and where the TSA or the new contracts raise questions of interpretation, the new entity own counsel should advise. The exit is delivered as part of our carve out and TSA separation service and the full carve out playbook.