Buying a Division? 6 Rules for the Transition Services Agreement Before the Seller's Systems Go Dark

When Lenovo bought IBM's personal computer business, announced in December 2004, it paid $1.25 billion and took on about $500 million in liabilities. What it did not get on day one was a company that could run on its own. The ThinkPad came with factories, engineers and customers, but the sales support, the financing and the warranty service still sat inside IBM. So the deal came with transitional arrangements: IBM kept providing those services for a period, and Lenovo was allowed to use the IBM brand on its PCs for up to five years while it built its own.

That is what buying a division looks like. You are not buying a company. You are buying a piece of one, and the piece has been living on the parent's payroll system, the parent's ERP, the parent's servers and the parent's bank accounts for years.

The contract that keeps that piece alive between closing and the day it can stand on its own is the Transition Services Agreement, and in most carve-outs it decides whether the first year is boring or a disaster.

Our group has grown by building and by buying for more than a century, across eight business lines. The lesson is always the same: the purchase agreement gets the lawyers' attention, and the TSA gets whatever time is left. These are the six rules I would apply before signing one.

What is a Transition Services Agreement?

A Transition Services Agreement (TSA) is a contract in which the seller keeps providing specific services to the business it sold, such as IT, accounting, payroll or logistics, for a fixed period after closing, so the buyer can keep operating while it builds or buys its own replacements.

It appears mostly in carve-outs, when a company sells a division, a brand or a plant rather than a standalone business. According to DealRoom's guide to TSAs, these agreements usually run 3 to 12 months, are priced at cost plus a small markup, and the longest workstreams are IT and infrastructure (around 12 months), ERP and accounting (around 10) and HR and payroll (around 9).

Those numbers sound short. They are not, if you have not started the replacement work on the day you sign.

Rule 1: Map every dependency before you sign, not after

The most common TSA failure is not a bad price. It is a service nobody listed.

A division uses dozens of shared things it does not think of as services: the parent's tax ID on the customs files, the group insurance policy, the shared warehouse lease, the EDI connection to the biggest retailer, the software licenses bought at group level, the person in head office who approves credit notes. If it is not in the TSA schedule, the seller has no obligation to keep providing it after closing.

Walk the business process by process, from order to cash and from hire to payroll, and write down every system, contract and person that sits outside the perimeter you are buying. That list becomes the TSA schedule. Anything you leave off, you will discover on a Monday morning.

This belongs in due diligence, not integration. If the seller cannot explain how the division would operate on day one without them, treat it like any other warning sign in our list of due diligence red flags.

Rule 2: Define each service like a contract, not a favor

"Seller will continue to provide finance support" is not a service. It is an argument waiting to happen.

Each line of the schedule should say:

• What exactly is provided: the process, the system, the volume.
• The service level: the same standard the business received in the twelve months before closing is the usual baseline.
• Who provides it: named teams, and ideally named people for critical services.
• How it is measured and reported: monthly, with a contact on each side.
• What happens if it fails: escalation path, credits or the right to step in.

The baseline matters most. The seller's incentive changes the day the money arrives. Before closing, the division is their business; after closing, it is a customer they are trying to get rid of. The contract has to hold the standard that goodwill no longer will.

Rule 3: Price it so both sides want it to end

Most TSAs are priced at the seller's cost plus a markup. DealRoom puts the usual markup at 5 to 10% and notes that fixed-fee TSAs typically cost 15 to 25% more than cost-plus, because the seller is absorbing the variability. For a typical carve-out, it estimates total TSA fees at 1 to 3% of deal value.

That money is not the real cost. The real cost is staying on the TSA too long. A cheap TSA with no price step-ups is a trap: it removes the pressure to build your own systems, and the seller's best people drift away while you wait.

What works better:

• Fair cost-plus pricing for the planned term.
• A step-up in price for each month of extension, agreed now, not when you are desperate.
• The right to end individual services early, with notice, and stop paying for them.

Rule 4: Make the exit plan part of the agreement

A TSA without an exit plan is a rental with no end date.

For each service, agree the target end date, the milestones that lead to it and what the seller must deliver at the end: data extracts in a usable format, documentation, knowledge transfer sessions, cooperation with your new provider. Data migration is where projects slip. DealRoom notes that cutover dates commonly slip two to three months past plan when dependencies on parent-company systems were not mapped during diligence.

Give the exit plan the same governance as the rest of the integration: one owner, a weekly review and a dashboard of services still running on the seller's systems. We set out that cadence in our week-by-week plan for the first 100 days after an acquisition; the TSA exit is one of its main workstreams.

Rule 5: Negotiate extensions and reverse services now

You will need more time on something. Plan for it while you still have leverage.

The purchase agreement is signed when the seller wants the deal most. That is the moment to secure:

• Extension rights on each critical service, at a pre-agreed price, at the buyer's option.
• Reverse TSA services if the seller will depend on you. In a carve-out, the division sometimes provides services back to the parent, such as a shared plant, a lab or a sales team. Those need terms too.
• Key-person commitments so the people who actually run the shared systems are not reassigned the week after closing.

Rule 6: Protect your data, your brand and your people during the transition

For months, the seller will still be running part of your business. That has consequences beyond cost.

• Data: the seller's systems will hold your customer lists, prices and payroll. Agree access controls, confidentiality and the deletion of your data at exit.
• Brand: if you keep using the seller's name, like Lenovo did with IBM, define where, how long and what happens to packaging and stock at the end.
• People: employees who move with the business may stay on the seller's benefit plans for months. Communicate clearly what changes and when, or you will lose people over a payroll mistake.

A TSA is a temporary partnership with someone whose interests are no longer yours, so write every protection you would want from a stranger. If retaining key people is a concern, combine it with the tools we covered in our rules for stay bonuses after an acquisition.

Key Takeaways

• A TSA keeps a carved-out business running on the seller's systems until the buyer can replace them; it is a bridge, not a home.
• The costliest mistake is a missing service: map every system, contract and person outside the perimeter before signing.
• Define each service with scope, a service level based on the pre-closing baseline, named providers and remedies.
• Price for exit: fair cost-plus for the planned term, with extension prices that rise each month.
• Every service needs an end date, milestones and a data handover agreed in the contract.
• Secure extension rights, reverse services and key-person commitments while you still have leverage.
• Protect data, brand use and employees as you would with any outside provider.

Frequently Asked Questions

Can you provide an example of a transition services agreement?

When Lenovo bought IBM's PC business in a deal announced in December 2004, IBM continued to support the business with services such as sales support, financing and warranty service for a transition period, and Lenovo could use the IBM brand for up to five years. More commonly, a seller keeps running the buyer's payroll, accounting system and IT network for 6 to 12 months after a division is sold.

How long does a transition services agreement usually last?

Most TSAs run between 3 and 12 months, according to DealRoom, with IT, ERP and payroll services taking the longest. Complex carve-outs and cross-border deals can need 18 to 24 months for some services. Each service should have its own end date rather than one date for the whole agreement.

How is a TSA priced?

Usually at the seller's actual cost plus a markup of around 5 to 10%. Fixed-fee pricing is simpler but tends to be more expensive. Good agreements add rising prices for extensions so that both parties have a reason to finish on time.

Who benefits from a transition services agreement?

Both sides. The buyer gets continuity, so customers and employees do not notice the change of owner. The seller gets a cleaner sale and a higher price, because a buyer will pay more for a business that will not stop working on day one.

What happens when a TSA ends before the buyer is ready?

The buyer either negotiates an extension, usually at a higher price, or runs without the service. That is why extension rights should be agreed when the purchase agreement is signed, not in the last month of the term.

Before You Buy a Division, Read the Fine Print That Keeps It Alive

A carve-out can be one of the best ways to buy a good business at a fair price, precisely because it is messy. The TSA is how you control that mess. If you want to see how a family group that has built and bought businesses since 1890 thinks about growth, learn more about Manzanos Enterprises.