Add-on acquisitions in private equity and who keeps the gain

What is an add-on acquisition in private equity?

An add-on acquisition is the purchase of a smaller company by a business that a private equity fund already owns, called the platform, usually at a lower multiple of earnings than the market puts on the platform. On paper the combined company is worth more than the two parts cost, and the sponsor keeps that gain only if integration works and the exit multiple holds.

Unlike the private equity deals that create a platform, an add-on is rarely bought by the fund itself. Bain & Company wrote in 2019 that the platform company usually makes the acquisitions, so it needs steady free cash flow to pay for them. A run of them is buy-and-build, which Bain defines as using a well-positioned platform to make at least four repeated add-on acquisitions of smaller companies. Bain counted the share of add-on deals that were at least the fourth by one platform: 21% in 2003, about 30% in its 2019 report (which its 2024 report recalls as more than 30%), and close to 50% in March 2024.

How common are add-on acquisitions in the US and Europe?

Add-ons were 72.9% of all US private equity buyouts by deal count in 2025, according to PitchBook data reported by the fintech-focused sell-side adviser Windsor Drake in August 2026: 5,351 add-ons against 1,985 platform buyouts, with a five-year average of 72.8%. PitchBook itself put the US share at 74.1% in its Annual US PE Breakdown, cited in a March 2025 article that does not state the year. A report cited in March 2025 cannot cover 2025, so the two figures are for different years and cannot be read as a rise or a fall. PitchBook put the US share at 68% in an October 2019 column drawing on its third-quarter 2019 report. The same March 2025 article put Europe at 66.7% of European buyouts in early 2025, below the US share, and no source we found gave a UK-only figure.

By value the share is much smaller. Windsor Drake reports that add-ons carried $350.7 billion, or 36.3%, of the $967.3 billion of US buyout value in 2025, so the count and the value point in opposite directions. In software the balance moved in 2026: PitchBook reported in June 2026 that add-ons, totalling roughly $18 billion, rose to about 45% of US software deal value in the first five months of the year, while platform deals fell to 41%.

Why does a platform pay a lower multiple for an add-on?

A platform pays a lower multiple because the market has historically valued larger companies above smaller ones, and the sponsor can have the add-on’s earnings valued at the platform’s multiple once they are inside it. Bain’s 2024 article calls this multiple arbitrage and says it brings down the average cost of acquisition, sweetening the return at exit.

The size of the gap varies by sector. In Bain’s 2019 account of Cressey & Company’s 2010 purchase of VetCor, small veterinary practices could be bought at a mid-single-digit multiple of EBITDA, against a mid-teens multiple for the platform. Bain adds that the arbitrage lets a sponsor justify paying a high price for the platform in the first place. PitchBook made a related point in 2019, when the median US buyout multiple had reached 12.9x: add-ons lower the average multiple a sponsor pays across its acquisitions over time.

Two things limit what a platform can pay. A price above its own multiple would leave no arbitrage, and the deal has to be funded, which Bain says usually falls to the platform’s free cash flow; the example below uses new debt.

How much value does an add-on create on paper, and how much does the seller get?

In the hypothetical example below, an add-on bought for $30 million creates $27 million of paper gain after costs, and an all-cash seller receives none of the gain left after the price. Amounts are in US dollars, every input is sourced or labelled as an example input, and the calculation is Unmatched Capital’s. The cash seller’s share is fixed by the price: at 8 times, or $40 million, the seller would take $10 million of the $30 million gross gap between 6 times and the platform’s 12 times, and $17 million of gain after costs would be left (Unmatched Capital’s calculation).

Input	Value	Basis
Platform	$25 million earnings (EBITDA), valued at 12 times: $300 million	The entry multiple in Bain’s 2024 example of a platform acquired several years earlier, which Bain says is probably worth less today; holding 12 times and the $25 million are our inputs
Platform debt	7 times earnings: $175 million, not repaid	The 7 turns are the debt in the same Bain example of a platform acquired several years earlier; no repayment is our assumption
Add-on	$5 million earnings, bought at 6 times: $30 million	Example inputs; Bain’s 2019 veterinary case describes small targets at mid-single-digit EBITDA multiples
Costs	$3 million of fees and integration, borrowed	Example input
Rollover	0%, 20% or 35% of the price	Advisers’ ranges for private equity sales generally, below
Rollover price	Platform equity of $125 million ($300 million less $175 million)	Our assumption

The paper gain is the add-on’s $5 million of earnings times the six turns between the 6 times paid and the platform’s 12 times: $30 million, or $27 million after the $3 million of costs.

Seller’s deal	Cash at close	Rolled stake	Stake worth at 12 times	Seller’s total	Share of the $27 million
All cash	$30.0 million	None	None	$30.0 million	0%
20% rollover	$24.0 million	4.58%	$7.24 million	$31.24 million	4.6%
35% rollover	$19.5 million	7.75%	$12.59 million	$32.09 million	7.7%

The rest goes to the platform’s existing shareholders: its private equity fund and its management. A rolled stake is a slice of the whole platform: the 20% seller’s $6.0 million joins $125 million of equity, buying 6 / 131, or 4.58%, and the borrowed $27 million of cash and costs leaves $158 million of equity at 12 times. The next table changes only the exit multiple; each roll shows the seller’s total, with the stake’s worth in brackets.

Exit multiple	Combined enterprise value	20% roll (rolled $6.0 million)	35% roll (rolled $10.5 million)	All-cash seller
12 times	$360 million	$31.24 million ($7.24 million)	$32.09 million ($12.59 million)	$30.0 million
10 times	$300 million	$28.49 million ($4.49 million)	$27.44 million ($7.94 million)	$30.0 million
8 times	$240 million	$25.74 million ($1.74 million)	$22.79 million ($3.29 million)	$30.0 million

Two turns off the multiple remove $60 million of value from the combined business, more than twice the $27 million gain. A rolled stake gets back the amount rolled at 11.1 times (the platform’s $300 million plus the $33 million paid for the add-on and its costs), before any return for the years the money is tied up; below that, a seller who rolled ends with less than the all-cash seller.

How does add-on practice differ between the US and Europe?

The sources show differences in the share of buyouts that are add-ons, in geography and in the tax on a rollover. PitchBook data put add-ons at 72.9% of US buyouts in 2025 (Windsor Drake, August 2026) and at 66.7% of European buyouts in early 2025 (PitchBook, March 2025). No source we found gave a UK-only share, or compared multiples, rollover sizes or earn-outs between markets, so how the price is set is not shown to differ.

	US	UK and the rest of Europe
Add-ons as a share of buyouts	72.9% in 2025 (PitchBook data, via Windsor Drake)	66.7% in early 2025 for Europe (PitchBook); no UK-only figure found
Rollover and tax	Only the equity actually rolled can defer tax; cash at closing is generally taxable (Porte Brown, June 2026)	UK management rollovers: a share exchange is not treated as a disposal for capital gains tax, and HMRC says its anti-avoidance rules should not be relevant where management have “a degree of choice” (HMRC Capital Gains Manual; Proskauer, June 2026)
Combining companies	Not found beyond Penningtons’s contrast	No general regime for merging UK companies, so a target is bought by a transfer of shares or assets (Penningtons, October 2019); the cross-border merger route Penningtons excepted has since been revoked (legislation.gov.uk)
Warranty insurance	Representations and warranties insurance, usually a buyer-led process (DLA Piper, January 2026; see the W&I page)	Warranty and indemnity insurance, which in an auction the seller can start and “staple” to the sale (DLA Piper, January 2026)

In Europe an add-on often crosses a border. Bain’s 2024 account of Cinven’s 2018 purchase of Partner in Pet Food, then Europe’s leading private-label pet food maker, describes five key acquisitions over three years, including a Swedish company in 2020 that gave access to the Nordic region and an Italian business that opened that market; revenue has tripled since Cinven bought it. That is one of Bain’s chosen cases, not a European average. Penningtons Manches Cooper, a law firm, wrote in 2019 that a UK target is usually bought by a transfer of its shares, which it called slightly more common, or of its assets, because UK law has no general regime for merging companies.

On rollover tax, Proskauer, a law firm, reported on 9 June 2026 that under section 135 of the Taxation of Chargeable Gains Act 1992 a share exchange is not treated as a disposal for capital gains tax, with the replacement shares stepping into the original shares’ position. HMRC’s Capital Gains Manual, updated for the anti-avoidance rule legislated in the Finance Act 2026, says that rule should not be relevant where private equity transactions allow management shareholders “a degree of choice in their investment”. Both concern management shareholders, a narrower group than every selling owner. This is general information, not tax advice.

None of this changes the arithmetic of the example, which is before tax. The tax on a rollover and the legal route for combining the companies are what differ by country.

What does the seller of an add-on usually receive?

The seller of an add-on can be offered a mix of cash at closing, rolled equity in the platform and sometimes an earn-out, and of the three, only rolled equity is a stake in the whole platform. No source read for this page gives rollover or earn-out figures for add-ons specifically. Advisers’ ranges for the rolled share in private equity sales generally differ. GHJ, an accounting and advisory firm, wrote in February 2021 that typical rolled amounts range from 20% to 35% in most transactions. Porte Brown, another accounting firm, wrote in June 2026 that rollovers in middle-market deals often fall in a 10% to 30% range. Both are practitioners’ estimates for private equity sales in general, not measured data and not add-on figures.

GHJ and Porte Brown both say the class of equity matters as well as the percentage. GHJ says buyers once asked for a preference ahead of the selling shareholder, and that sellers have since secured equity that is pari passu with the private equity group. Porte Brown warns that a seller holding junior common equity beside a sponsor’s preferred equity may own something worth less than the headline price. GHJ adds that if the business underperforms, the rolled piece “may collapse all the way to nothing.” Porte Brown, writing for US sellers, also says only the equity actually rolled can defer tax, that cash taken at closing is generally taxable, and that tax-deferred does not mean tax-free. The tax treatment of a rollover differs by country, as the section above shows for the UK.

An earn-out is a payment that depends on the add-on’s later performance. SRS Acquiom, which describes itself as shareholder representative on over 4,600 deals, reports in an earn-out summary that 24% of 2025 private-target deals (excluding life sciences) included one, up from 19% in 2014, and in its lower middle-market report that 35% of the smallest lower middle-market deals (up to $25 million) did. Neither figure is specific to add-ons: the lower middle-market deals, those with closing payments of $50 million or less, are drawn from SRS Acquiom’s dataset of more than 4,400 private-target transactions, and private equity buyers took part in only 11% of them. Across SRS Acquiom’s own deals with an earn-out, closer to one dollar in five gets paid; its 2026 study puts the achievement rate, excluding life sciences, at 21 cents on the dollar. That is an aggregate across many deals, not a forecast for any one earn-out.

The indemnity a buyer asks a seller to stand behind is a separate negotiation, covered in warranty and indemnity insurance.

What happens to the business and its staff after an add-on?

What happens depends on the platform’s thesis, and no readable dataset measures staff outcomes for add-ons specifically. The nearest measured evidence covers buyouts in general. A National Bureau of Economic Research working paper by Steven Davis, John Haltiwanger and co-authors, covering US buyouts from 1980 to 2013 and revised in April 2024, found that employment shrinks 12% over two years after buyouts of publicly listed firms and expands 15% after buyouts of privately held firms, relative to control firms. The paper measures buyout targets. In a buy-and-build the target is the platform, so add-ons bought later may appear inside the platform’s job growth, and the abstract does not say how they are counted. These figures cannot be read as add-on outcomes.

The standard case for cost savings includes cuts. Wall Street Prep, a finance training firm, lists among cost savings the elimination of overlapping workforce functions and reduced headcount (February 2024). Bain’s accounts of successful buy-and-builds describe other routes. At Caliber Collision, Bain wrote in 2024, a “tiger team” trains each new shop’s staff while replacing its IT systems; in VetCor, Bain wrote in 2019, the plan let clinics keep their local identities and left resident veterinarians in control of medical decisions. Those are Bain’s chosen success stories, not averages. Our reading is that an add-on which duplicates the platform’s back office has more to cut than one that brings a new region or product.

When does the add-on arbitrage fail?

The arbitrage fails when the platform’s multiple falls, when rival buyers bid targets up, when integration costs more than planned, or when debt absorbs the gain. Bain’s analysis of 44 buy-and-build deals made between 2010 and 2019 found that those depending on multiple arbitrage alone averaged a 1.4x multiple on invested capital, against 2.2x for those with a strategic rationale driving organic growth or margin improvement. Bain does not name the deals or state its method for classifying them.

Higher rates change the sum. Bain’s example platform, bought at 12 times earnings with 7 turns of variable-rate debt, faces financing costs approximately double what investors modelled in 2020, and Bain wrote that higher rates are putting downward pressure on asset prices. In our example two turns off the exit multiple cost $60 million, of which $50 million falls on the platform’s own earnings and would be lost even without the add-on. Rival buyers erode the arbitrage from the other side: Bain warned in 2019 that too much demand for too few targets drives up prices, and pointed to US funeral homes, where consolidators picked off the attractive midsize targets.

A years-long reclamation process can cut deeply into ROI.

Bain’s cautionary case was Aurora Foods, which bought brands from 1996, strayed from its core and filed for bankruptcy in 2003. PitchBook found in 2019 that platforms making add-ons took about a year longer to exit than non-acquisitive ones. When the platform is sold to another sponsor, that sale is a secondary buyout, and the new owner prices the combined business at its own multiple.

What this means if a platform approaches your company

An owner approached by a platform is being offered a price built on the platform’s multiple, which was 12 times earnings in Bain’s 2024 example of a typical platform bought several years earlier, so the useful things to establish are how big the gap is, how many other buyers exist, and what the rolled equity is worth if the exit goes badly. Unmatched Capital’s analysis is that a cash seller’s share of the arbitrage is fixed by the price at closing, which rises only if buyers compete and, unless the platform expects cost savings or growth from the add-on, cannot sensibly exceed the platform’s own multiple, and that a rollover pays for owning part of the platform, which is a different thing from being paid for the business that was sold.

The strongest counterargument is that the platform carries the debt, the integration and the exit risk that make the gain possible, and that a rolled stake also shares in later add-ons and growth, which our single add-on leaves out. Both points are fair, and they justify the platform keeping most of the gain. They do not justify a price below what competing buyers would pay, because the platform carries that risk whatever it pays, and a bidder facing rivals has to pay up to the point where what remains of the gain only just compensates it. A seller who wants a share of later add-ons and growth can have it only by rolling, which brings the exit risk shown in the second table.

A seller approached by a platform can ask five questions before accepting:

1. What multiple is the offer, and what multiple was the platform bought at?
2. How many other consolidators are buying in this sector?
3. If I roll, what class of equity is it, at what price, what protects it from dilution by later add-ons, what happens to it if I leave or the sponsor sells, and how is the rollover taxed where I live?
4. How much of the price is an earn-out, on what measure, and who runs the business while it is measured?
5. What is the rolled stake worth if the exit multiple is two turns lower?

An LP or adviser reading a fund’s buy-and-build case can ask the same of the manager: whether the thesis rests on arbitrage alone, which is the case Bain’s 1.4x against 2.2x describes. This is general information, not advice.