The TSA is the contract. The SLA is the teeth.

Section 01

Two different things, one document. Contract and clause.

A Transition Services Agreement is the contract. It defines which services the seller provides after closing, for how long, at what price, under what governance, and how the buyer exits each one. It is a complete legal instrument negotiated against the deal timeline. The service-level agreement is one part of that instrument: the section that says how well each service has to perform.

Outside of carve-outs the term SLA often refers to a standalone document, which is where the confusion starts. Inside a TSA the SLA is not a separate contract. It is the set of measurable commitments attached to each service in the catalog. Response times, uptime, processing windows, accuracy thresholds, and the remedies that apply when those commitments are missed all live in the SLA portion of the TSA.

The practical point for a buyer is that a TSA without real SLAs is a promise to provide services with no defined quality and no consequence for poor delivery. The seller is winding down a business it sold, its best people are leaving, and its incentive to perform is low. The SLA is the mechanism that converts a vague obligation to help into a measurable obligation to perform.

Section 02

What a service level actually specifies. Measurable, not aspirational.

A usable service level names a metric, a target, and a measurement method. For a payroll service that might be pay runs completed accurately by an agreed cut off each cycle. For an IT service it might be system availability during business hours and a resolution time for priority incidents. The metric has to be observable by both parties, because a service level no one can measure is a service level no one can enforce.

Targets should reflect what the buyer needs to run the business, not what the seller finds convenient. Sellers often propose service levels that match their own internal averages, which were achieved when the function had full executive support and the business was not being separated. Buyers should set targets against operational necessity and require the seller to commit to them in writing for the duration of each service.

Each service level also needs a measurement period and a reporting cadence. Monthly reporting against agreed metrics gives the governance committee something concrete to manage. Without a defined measurement period, disputes about performance become arguments about memory. With one, performance becomes a number both sides can see, and the conversation shifts from blame to remedy.

Section 03

Service credits. The remedy with teeth.

A service credit is the money the buyer gets back when a service level is missed. It is the enforcement mechanism that gives the SLA its force. Without credits, a missed service level is a complaint. With credits, it is a deduction from the next invoice, which gives the seller a financial reason to perform even while it is winding the function down.

Credit structures usually scale with the severity of the miss. A small breach of a target might return a few percent of the monthly service fee, while a sustained or severe failure returns more, often capped at a quarter to half of the monthly fee for that service. The cap protects the seller from unlimited exposure, but the buyer should resist caps so low that the seller can underperform cheaply and treat the credit as a cost of doing business.

Credits are a remedy, not a substitute for performance. A buyer whose payroll fails does not want a credit, it wants the payroll run. So the SLA should pair credits with escalation and, for critical services, with step in or termination rights when failures persist. The credit handles the routine miss. The escalation handles the failure that threatens the business.

One refinement matters: the credit should be automatic, not earned by complaint. If the buyer has to detect the miss, document it, and request the credit, many misses go uncollected because the buyer is busy running a transition. A well drafted SLA makes the seller report its own performance against the metric each period and apply the credit at source, with the buyer auditing rather than chasing. Automatic credits convert the service level from a clause the buyer must enforce into a discipline the seller must maintain.

Section 04

Why buyers underweight SLAs. And pay for it later.

In the rush to close, SLAs are often the last thing negotiated and the first thing conceded. The commercial terms get the attention, the service levels get a cursory pass, and the buyer signs a TSA with service commitments that look fine on paper and provide no protection in practice. The cost shows up months later when a service degrades and the buyer discovers it has no measurable standard to point to.

The seller has every reason to keep service levels soft. Soft service levels mean less exposure to credits and less obligation to staff the function properly during wind down. A buyer who does not pressure test the SLAs during negotiation inherits whatever the seller proposed, which is calibrated for the seller’s convenience and not the buyer’s continuity.

The fix is to treat service levels as core deal terms, negotiated before signing with the same seriousness as price. Each critical service should have a metric, a target set to business need, a measurement method, a reporting cadence, and a credit that actually bites. This is exactly the work a buyer-side review performs, and it is far cheaper to do before signing than to litigate after a failure.

Section 05

How the SLA supports the exit. Performance buys time.

Strong service levels do more than protect daily operations. They protect the migration. A buyer can only execute a clean, sequenced exit if the underlying services keep performing while the migration runs. If a service degrades during transition, the migration plan slips, the TSA extends, and the buyer pays extension fees on a service it was trying to leave.

Reliable service levels give the buyer a stable platform to migrate from. The team can plan cutovers, test in parallel, and retire services on schedule because the seller is meeting its commitments throughout. In that sense the SLA is not just a quality control. It is a precondition for the exit ramp to function as designed.

All of this work, the service catalog, the service levels, the credits, and the exit ramp, is negotiated together before signing through a disciplined TSA exit approach. The SLA is the part that makes the rest enforceable. A buyer who insists on real service levels is buying both daily protection and a credible path off the seller’s systems.

FAQ

TSA versus SLA questions buyers ask.

Is an SLA the same as a TSA?

No. A TSA is the full contract that governs transition services after a carve-out. An SLA is the performance standard inside the TSA that defines how well each service must be delivered and what the buyer is owed when it is not.

What happens if the seller misses an SLA?

If the TSA is drafted well, the buyer earns a service credit, a deduction from the service fee scaled to the severity of the miss. For repeated or severe failures the buyer should also have escalation, step in, or termination rights, because a credit alone does not fix a critical service that keeps failing.

Who sets the service-level targets?

They are negotiated. Sellers tend to propose targets that match their internal averages. Buyers should set targets against what the business actually needs to operate and migrate, and require the seller to commit to those targets for the full term of each service.

Are service credits capped?

Usually yes. Most TSAs cap monthly credits at a quarter to half of the monthly service fee to limit seller exposure. Buyers should make sure the cap is high enough that underperformance is genuinely costly to the seller rather than a cheap cost of doing business.

Related Reading

More on the TSA cluster.