Procurement in private equity value creation

Procurement is one of the few value creation levers a sponsor controls directly and can start within weeks of close. It converts into EBITDA, releases working capital, and removes risk a buyer would otherwise discount at exit. Finding the money is the first job. Making sure it is still there, and still provable, at exit is the second, and it is the one most programmes lose.

Why the return now has to be earned operationally

Multiple expansion and cheap debt did most of the work in the last cycle and are not available now. McKinsey attributes 59% of returns on 2010 to 2022 deals to leverage and multiple expansion. Bain puts the operating consequence plainly: a deal underwritten in the 2010s needed roughly 5% annual EBITDA growth to reach a benchmark 2.5x return over five years, and the same deal today needs closer to 10% to 12%.

The exit data has already moved. Alvarez & Marsal’s analysis of 240 Western European exits found margin expansion accounted for 51% of EBITDA growth for businesses exited in 2025, against 21.5% before 2023. Fifty-eight per cent of firms now deploy operational resource inside the first 100 days, double the prior year.

What procurement should deliver in a portfolio company

Cost is the primary objective and there is no reason to dress it up as something else. But a sourcing event run properly produces five outcomes, and in a leveraged structure the second and the fifth are frequently worth more than the first.

Most programmes deliver the first and leave the other four largely untouched. That is why two owners can run procurement on the same business and report very different numbers, and it is worth establishing which of the five a plan is actually underwriting before it is signed.

EBITDA, multiplied at exit. A krone of cost removed is a krone of EBITDA, and at a 10x multiple it is ten kroner of enterprise value. Nothing else available to a sponsor converts at that ratio on that timescale.

Cash, which does not show in EBITDA. Payment terms, delivery frequency, minimum order quantities, consignment and inventory held on the supplier’s balance sheet rather than yours. Released working capital pays down debt or funds a bolt-on without going back to the fund.

Own versus lease belongs here, and it is a category decision rather than a treasury one. Take tools and equipment in a construction business: a category optimisation on that spend has to answer how the business should acquire it going forward, buy or lease or rent by the job, by asset class and by utilisation. That question is commercial, technical and process at once. Once the future model is settled the installed base is a legacy of the old one, and sale and leaseback is simply how you migrate what you already own onto the model you have chosen. Selling down slow-moving stock follows the same logic, because what you bought and never consumed tells you both what to sell and what to stop ordering.

This is not make versus buy. That asks whether the business should produce something itself, an operating model decision on a multi-year horizon that belongs elsewhere. Own versus lease asks how you acquire something you are buying either way.

Better cost. A specification that matches what the business sells today rather than a decade ago. Standardisation that removes variants nobody asked for. Sometimes a higher specification, where it lowers lifetime cost or reduces failures customers notice. At exit this shows up as margin quality and customer retention, which is what a buyer underwrites.

Risk a buyer will otherwise discount. A single unqualified source, a contract with no exit, a price with no index or cap. Each is a discount at exit applied by a buyer who found it in diligence, and the discount is larger than the cost of fixing it during the hold.

A repeatable capability, which is the one that compounds across a portfolio. This matters more to a sponsor than to any single company. In a buy-and-build platform, a procurement function that can integrate a bolt-on onto agreed terms in weeks rather than quarters is part of the acquisition engine. It shortens time from close to synergy on every subsequent deal, it makes the platform a more credible acquirer, and it is a capability the next owner is buying rather than a saving they have to take on trust. Build it once in the platform company and it works on every deal after it.

What the numbers look like

Take a Nordic portfolio company with NOK 1.5bn revenue and NOK 150m EBITDA. Assume external spend at 45% of revenue, so NOK 675m, and 75% of that addressable, so NOK 506m. An expected range of 7% to 12% on that base gives NOK 35m to 61m, an uplift of roughly a quarter to two-fifths on EBITDA. At a 10x multiple that is NOK 354m to 608m of enterprise value.

Three things about that figure. It covers the cost line only, so working capital, better cost, risk and capability sit outside it. It is un-risked, carrying no deduction for cost to achieve or for the erosion described below, and it describes value signed off against an agreed baseline rather than cash that has left the business. And the input assumptions do the heavy lifting: external spend as a share of revenue varies widely by sector, and the figures quoted across consultancy material carry no stated methodology at all. Our 7% to 12% comes from our own delivery experience, not from a benchmark study. Run the arithmetic on the target’s own numbers or do not run it.

What procurement is for at each stage of the deal

Stage	What procurement is for	The thing to get right
Pre-deal	Sizing the opportunity as a range rather than a number. Look wider than categories: payment terms, inventory and supplier concentration are all sizeable and visible earlier than category detail	Be honest about what a data room cannot tell you. A top-twenty supplier list and a trial balance show concentration and fragmentation. They cannot validate a baseline or reveal a specification. Underwrite the conservative end
First 100 days	A diagnostic, a savings rulebook agreed with the CFO, then two or three sourcing events that can be evidenced quickly	The rulebook beats the quick wins. Agreeing what counts as a saving, against which baseline, before anyone negotiates, prevents the argument that arrives in year three. Run the payment terms review in the same window, along with the quick asset moves: sale and leaseback where the future model is lease, and selling down slow-moving stock and idle equipment
Hold period	Timing constrains less than people assume. In a business of this size maximum signed-off value is reachable within roughly eighteen months, with full run-rate realised about six months after that	So even a three-year hold accommodates the whole programme, and the hold period rarely decides which approach you run. What a longer hold changes is the payback on capability. Nordic holds have gone from 4.3 years in 2020 to 6.2, with 40% of exits beyond seven, and at that length a function that can repeat the exercise unaided is worth more than the first wave, because the wave erodes and the capability does not
Carve-out	The most under-priced procurement risk in the market. The business loses parent framework agreements on day one and has no purchasing history of its own	BCG puts one-time separation costs at 1% to 5% of divested revenue, up to 13% in complex cases, with transitional service agreements running three to twenty-four months. Do not inherit the parent’s contracts unexamined. Note also that those benchmarks predate current tooling: faster execution and tighter tracking should compress both the one-time cost and the leakage that follows it, though nobody has published a figure on that yet
Buy-and-build	The Nordic case: 82% of Nordic deals in 2025 were add-ons. Aggregating spend across the platform, and building the capability to do it again on the next one	Specification harmonisation is the work and the negotiation is the easy part, but the hardest part is neither. It is mobilising ownership and leadership across the companies behind one procurement agenda. That is what closes the gap between signed off and realised, it is what makes the capability compound into the next acquisition, and it is what persuades a buyer the savings will stick. Where portfolio companies compete with one another, joint purchasing raises competition law questions under EU and EEA rules; get counsel’s view before the first joint tender

Making savings stand up to a quality of earnings review

Two bodies of professional literature have never been joined up. One covers procurement savings tracking, the other covers quality of earnings, add-backs and run-rate EBITDA. At exit they collide.

A buyer’s quality of earnings provider does not accept a saving because procurement says it happened. It asks what the baseline was, whether the volume assumption held, and whether the price is actually flowing. A saving negotiated in month thirty and presented as run-rate EBITDA at exit will be tested, and the discount applied to an untested claim is larger than the one applied to a smaller, evidenced claim.

We work to a five-stage scale, and it exists to survive that conversation.

The logic underneath it is what makes it useful in diligence: each stage is one further party committing to the number. A buyer does not ask how confident you are. They ask who has signed.

Stage	The test	What evidences it	Who has committed
Estimated	An outside-in percentage applied to a spend base	A benchmark. No supplier named	Nobody
Assessed	The opportunity sized bottom up from actual spend, the supply market and the specification	Our own analysis across commercial, technical and process levers, before anyone is approached	Us
Identified	A specific change is agreed and priced, whether that is a supplier price, a specification, a volume or a process	An agreed position, not yet approved internally	The supplier, or the function that has to deliver it
Signed off	The calculation is approved against the agreed baseline	Run rate, effective date, budget holder and finance both accepting	The client
Realised	The invoice arrives at the agreed cost, or against the agreed specification	Invoices reflecting the change, at the volume assumed	The ledger

Only the last two survive a buyer’s scrutiny. The discipline worth building from month one is documenting the fourth, because it cannot be reconstructed at exit.

Two things the scale is easy to get wrong.

Every stage measures total cost of ownership, not price. A specification change, a reduction in variants, a process that removes handling, an inventory level that comes down: each is converted to kroner and sits on the same ladder as a negotiated discount. A scale that only tracks price movements makes two of the three lever families invisible, and the number you report at exit is then smaller than the value you actually created.

The gap between stages four and five is where deals lose their value, and closing it is organisational work. A signed-off saving is paper money. It becomes real only when the specification has actually changed in the systems and standards people order from, the chosen supplier is the one actually used, the right item is bought rather than a near equivalent, and people buy it in the way the new agreement assumes. Adjusting the budget so the money cannot quietly be respent is one of the vehicles for that, and it belongs immediately after sign-off as implementation preparation, not as a test of whether realisation happened.

Measure the gap on purpose. Track what share of signed-off value has reached the P&L, initiative by initiative, with the shortfall logged and an action against each one. A programme that reports signed-off value and never reports leakage is telling you half the story, and it is the half a buyer discounts.

Why the last owner did not already take this

Most Nordic targets have already had a procurement programme run at them, so expect to be asked why the last owner did not take this.

The honest answer is rarely that the previous programme was bad. In our experience it was usually run competently all the way to sign-off. What went wrong came afterwards, and it was almost always the same thing: the organisation was never mobilised behind it. Specifications were not changed in the places people actually order from. The selected suppliers were not the ones used. Near-equivalents were bought instead of the agreed item, or the agreed item was bought in a way the new terms did not cover. Nobody owned the gap, so nobody closed it.

Which means the question worth putting to a target is not what the last programme identified. It is what it realised, and whether anyone measured the difference. Very few can answer the second part, and that is itself the finding.

It is also where we put our weight. Sign off as much as is genuinely available, and design for realisation from the start by working out what has to happen in the organisation during the process rather than after it. Run that way it stops being a paper exercise aimed at signed-off value and becomes a considered one aimed at realised value, which is the only number a buyer pays for.

These are structural conditions rather than failures of the people involved, and each is diagnosable in a fortnight.