Method
How to assess a carve-out before signing.
Before an investment committee approves a carve-out, it should have five questions answered in numbers, not narrative. Each answer should be comparable across deals and specific enough to change the price, the terms or the plan.
Question 1
Can the separation be done on the current plan?
A carve-out can be attractive and still be impractical to separate in the time and budget the deal assumes.
Evidence to use
How entangled the ERP and IT estate is, shared sites and people, contracts that need consent, licences and regulatory approvals, and the timetable to close.
What the IC should see
A clear feasibility rating, and the specific issues that drive it.
In CarveOutIQ
The Execution Feasibility Index: a 0–100 score of how achievable the separation is on the current plan, banded Contained, Elevated or Severe, with the red flags behind it.
Question 2
How dependent will the business be on the seller after close?
Transitional services keep the business running, but every month of dependence costs money and every overrun costs more.
Evidence to use
TSA scope by function, the proposed duration and pricing, and how hard each service will be to replace.
What the IC should see
The cost of the TSA as agreed, the cost if it overruns, and where exit will be hardest.
In CarveOutIQ
TSA exposure: contracted TSA cost, the cost of overrunning, and exit complexity by function, shown as a heatmap.
Question 3
Will the business be ready to operate on Day 1?
A business that cannot pay people, close its books or serve customers on its own on Day 1 puts the investment case at risk immediately.
Evidence to use
Readiness of finance, IT, people, legal and operations: the systems, processes and people each needs on its own.
What the IC should see
Readiness as a percentage against a target, and the gaps that must close before completion.
In CarveOutIQ
The Operational Readiness Index: Day 1 readiness as a percentage against a 75% target, across finance, IT, people, legal and operations.
Question 4
What will the separation cost, one-off?
One-off separation cost is cash the fund will spend that rarely appears clearly in the investment case.
Evidence to use
IT separation, standing up standalone functions, programme and adviser costs, and transition costs.
What the IC should see
A one-off cost in £m, shown separately from recurring effects on EBITDA.
In CarveOutIQ
One-off separation cost in £m, as part of combined separation exposure.
Question 5
How much EBITDA will leak, and how much of the plan can the deal deliver?
Replacing shared services, running duplicate costs and absorbing disruption all reduce the EBITDA the investment case is built on.
Evidence to use
Standalone running costs, duplicate and transition costs, the risk of TSA overruns, and the likely disruption to the business.
What the IC should see
EBITDA leakage in £m, and the share of plan EBITDA the deal can realistically deliver.
In CarveOutIQ
EBITDA leakage in £m and Value Realisation Potential: the share of plan EBITDA the deal can deliver after separation, as a percentage and in £m.
Worked calculation
Project Atlas, from inputs to the IC figures
Atlas is an illustrative healthcare carve-out in diligence: £300m revenue, £45m plan EBITDA, £240m enterprise value, a fully shared ERP, a 12-month TSA and 14 legal entities. This is how its figures are built.
1
Score the nine separation dimensions
TSA dependency, IT, finance, legal entity, people, contracts, operations, Day 1 governance and leadership, each scored 1 to 5 from the diligence in hand against written scoring anchors. Atlas scores 5 on IT separation and 4 on governance, TSA, contracts and leadership.
2
Day 1 readiness: Operational Readiness Index 31%
Readiness starts at 100% and is reduced by separation complexity, a compressed timetable, ERP entanglement, geographic spread and regulatory dependencies. Regulated perimeters carry a ceiling until licences and registrations transfer. Atlas loses most through IT complexity, the fully shared ERP and its regulated healthcare perimeter.
3
Feasibility: Execution Feasibility Index 46 / 100
Combines Day 1 readiness (the largest component), TSA length, input completeness and one-off cost relative to enterprise value. Banded Contained (75 and above), Elevated (50 to 74) or Severe (below 50). Atlas is Severe.
4
Exposure: £32.0m one-off cost and £17.5m EBITDA leakage
One-off cost covers IT, people, finance, legal and transition work, scaled for the deal's size and complexity. Leakage covers revenue disruption, TSA overrun, duplicate running costs and the cost of delay to Day 1, modelled for the separation period. Together, £49.5m: shown for scale, not as a change to enterprise value.
5
Value Realisation Potential: 36%
£45.0m plan EBITDA, less £17.5m leakage, less £10.7m one-off cost per year (£32.0m over three years) leaves £16.8m realisable, or 37.4% of plan. That is discounted by input completeness (96% for Atlas): 37.4% × 0.96 = 36%.
6
Value each issue
Each red flag is re-run as if addressed, to show the exposure it removes. Addressing Atlas's nine addressable issues together lifts VRP to 54% and readiness to 48%, and cuts combined exposure by £12.3m.
The weights and calibration constants behind each step are shared with clients under NDA. Every figure in the app can be traced back to the inputs that drive it.
What the method rests on
Practitioner judgement, being tested against outcomes
It is important to be clear about what is calibrated and what is not.
- Practitioner-set today. The scoring anchors, the weights, the thresholds (75 and 50 for the bands, the 75% readiness target) and the cost and leakage rates are set from separation experience across chemicals, FMCG, healthcare, telecoms and retail carve-outs. They are not yet statistically validated.
- Validation under way. A study is back-testing the method against public disclosures of completed carve-outs, using only information that was available at the decision date. Every scored pilot deal also records its actual outcome, so the method can be tested against results.
- Correlated by design. The three indices share inputs and are meant to move together: three views of one execution model, not three independent opinions.
- Deterministic. A fixed set of rules and calculations. The same inputs always give the same figures, and no AI model processes deal data.
The 75% readiness target
Why 75%, and what it is not
At around 75% readiness, the functions a business needs on Day 1 (paying people, closing the books, invoicing and serving customers, running core IT) are either standalone or covered by an agreed TSA with a tested cutover. Below that, Day 1 typically depends on workarounds, and TSA extensions and delays become likely.
The target is a practitioner benchmark, not a statistical cut-off. A deal below 75% is not unsellable: it means the gaps should be closed, priced or written into conditions before signing. The validation study will test where the line should sit.
From answers to decisions
- Combine the exposure. One-off cost plus EBITDA leakage gives one figure for what is at stake if the issues are not addressed. Leakage is the modelled total for the separation period. The combined figure shows scale; it is not an adjustment to enterprise value, because one part is cash and the other is EBITDA.
- Value each issue. For every red flag, show how much exposure addressing it removes, so the team knows where to spend effort first.
- Turn issues into conditions. Set out what must be agreed before signing or close, worded for the SPA and TSA negotiation.
- Test the range. Run Low, Base and High scenarios, and change assumptions such as TSA duration to see the effect.
- Use the same method on every deal, so a pipeline can be compared and ranked on execution risk.
This does not replace operational due diligence, the data room or the financial model. It turns what they find into figures an investment committee can act on, early enough to change the deal. See it applied to an illustrative £240m carve-out in the worked example.