Carve-outs and divestitures done right: a CFO's execution playbook

The pressure to divest is real and, in 2026, it is intensifying. Activist shareholders are more organised, interest rates remain elevated enough to punish sprawling conglomerates, and private equity buyers have dry powder ready for focused assets. When GE completed its three-way split into GE Aerospace, GE Vernova, and GE HealthCare, the process took years and still produced hundreds of millions in separation costs. That was a company with significant resources dedicated to the effort. Most organisations attempting a carve-out have far less runway and far fewer people who have done it before.

The most common mistake is treating a divestiture as a deal problem rather than an operations problem. Once the term sheet is signed, the real work begins: separating IT infrastructure, untangling shared service agreements, splitting pension obligations, and telling customers who their new counterparty is. CFOs who underestimate this phase routinely see transaction value eroded by transition service agreement (TSA) fees, delayed separations, and a distracted core business.

The execution sequence

Step 1: build the separation management office before you announce

The moment a divestiture is even probable, stand up a dedicated separation management office (SMO) with a full-time leader who reports directly to the CFO. This is not a project manager role. You need someone who understands both financial mechanics and operational interdependencies. Staff it with representatives from IT, HR, legal, treasury, and the business unit being divested. The SMO's first output is a dependency mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.View full definition →: every shared system, shared contract, shared employee, and shared facility. Nothing clarifies scope faster than this exercise, and it will surface surprises you cannot afford to discover in due diligence.

Step 2: price the TSA honestly and plan its exit

TSAs are the mechanism that keeps the divested entity functioning post-close while it builds its own capabilities. They are also where deals quietly bleed value. Buyers negotiate hard on purchase price and then absorb elevated TSA fees for 18 to 24 months because no one modelled the exit properly. As the seller, you want TSAs to be short, priced at full cost recovery plus a reasonable margin, and structured with hard exit dates. As the buyer, you want the opposite. Either way, the critical discipline is building a TSA exit plan before close, not after. Specify the target state, identify the IT builds or vendor contracts required, and assign accountability for hitting each milestone.

Step 3: clean the financials early and keep them clean

Buyers and their advisors will stress-test standalone financials. If the carve-out has been using shared overhead allocations that do not reflect economic reality, restatements will be required, and they will slow due diligence. Start by producing two to three years of standalone financials on a fully allocated, arms-length basis. Identify costs that will not follow the business (a portion of corporate HQ, for example) and costs that will need to be added (a standalone CFO, a separate audit, D&O insurance). Investors and sponsors look at EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition → adjustments with considerable scepticism, so the cleaner and better-supported the numbers, the less haircut you take on valuation.

Step 4: manage the workforce with precision

Employee retention during a carve-out is chronically underestimated. The people with the deepest institutional knowledge are exactly the ones with the most options. Retention agreements tied to deal close and the first anniversary of separation are standard, but the structure matters. Cash-only retention bonuses paid entirely at close create a cliff: the employee walks the day after the money lands. Staggered payments, with a portion at close and a portion 12 months later, align incentives better. For the employees who will move to the new entity, clarity on benefits continuation, pension treatment, and equity stakes in the separated business needs to come early. Uncertainty is the accelerant for attrition.

Step 5: communicate with customers and suppliers before rumours do

In a carve-out, customers often discover the change from a news alert or a supplier portal update. That is a relationship risk. Identify the top 20 customers by revenue in the divested business and have a prepared communication sequence ready to execute on announcement day. The message should be specific: who their account manager will be, whether their contracts are being novated or assigned, and what the service continuity plan is. For suppliers, prioritise any contract that requires consent to assignment, because those become negotiating leverage for the counterparty if surfaced late.

Where this goes wrong

The most destructive failure mode is what practitioners call "stranded cost": overhead that was allocated to the divested unit but cannot actually be removed from the seller's cost base. GE's experience, and similar ones at Siemens and Philips during their respective healthcare separations, showed that stranded costs of 15 to 25 percent of allocated overhead are common. Model this explicitly before signing the deal, not after.

The second failure mode is an IT separation that takes twice as long as projected. The dependency map from Step 1 is only as good as the quality of your enterprise architecture documentation, which in most companies is incomplete. Build in a contingency of at least six months beyond your initial IT separation timeline.

A third, less discussed failure is valuation leakage from a distracted management team in the core business. When the leadership of the retained entity spends 40 percent of their time on separation activities, the core business performance softens. The SMO structure exists precisely to insulate the operating teams.

Quick wins to start this week

• Pull the list of all shared IT systems and contracts that touch the business unit in question. Even a rough inventory will surface the first critical dependencies.
• Ask the CFO of the unit to model fully standalone financials for the last two fiscal years, using arms-length cost allocations. Do not wait for a formal process.
• Identify which employees in the unit are flight risks and check whether any retention mechanisms are currently in place.
• Have legal flag every material contract that contains a change of control or assignment clause.

Execution quality is what separates divestitures that deliver target returns from those that erode them. The strategic rationale is rarely the problem. The separation mechanics are.