It is easy to look back at a profitable buyout and say the deal “worked.” It is a different exercise to say why — whether the gain came from paying down acquisition debt, growing the business, or simply selling at a richer multiple than the one paid on entry. Those three levers respond to completely different things, and only one of them is inside a buyer’s control after closing.
Key takeaways
Canada’s private capital market moved $12.7 billion across 252 private equity deals in the first half of 2026, according to CVCA — though four privatization transactions alone accounted for $7.25 billion of that, or 57% of the total capital deployed. That detail matters for a small buyout specifically: strip out four large take-privates and the remaining 248 deals averaged well under $25 million each. The headline number describes a market dominated by a handful of large transactions; the mechanics that decide whether a small buyout actually returns money are the same regardless of deal size, and they come down to three separate, independently moving levers.
Strip a buyout’s outcome down to its arithmetic and there are only three places the gain can have come from: the acquisition debt got paid down, so more of the exit value belongs to equity; the business earned more at exit than it did at entry, so the same multiple is now applied to a bigger number; or the multiple itself moved, so the same earnings are now worth more — or less — to a buyer. Every buyout’s return is some combination of the three, and they respond to entirely different things. A buyer who cannot say which lever produced the return cannot tell whether the deal actually worked, or whether it was carried by a market that happened to be generous on the way out.
Debt paydown is the most mechanical of the three: every dollar of principal repaid during the hold is a dollar that moves from the lender’s side of the balance sheet to the equity holder’s. It is also the lever most shaped by how the deal was financed in the first place, and Canadian financing rules narrow the available paths more than buyers often expect. The Canada Small Business Financing Program’s own FAQ is explicit that “you cannot use a loan to finance items such as share purchases or assets that a holding company acquires” — the programme only reaches an asset purchase, and even there the per-borrower loan maximum is $1.15 million, split across real property, equipment and leasehold improvements, and a smaller working-capital allowance. A vendor who wants share-sale tax treatment and a buyer who wants CSBFP-eligible debt are, structurally, pulling in different directions, and that tension shapes how much of the capital stack can actually amortize on favourable terms.
Where a seller finances part of the price directly instead, the Income Tax Act’s s. 40(1)(a)(iii) capital gains reserve caps an ordinary arm’s-length vendor take-back at a five-year spread — “1/5” of the gain multiplied by the number of years remaining out of four. That five-year outside limit on the reserve tends to set the natural repayment horizon for a VTB note in the first place: a seller rarely wants loan terms that outrun the tax deferral built around them, which means the debt-paydown lever in a seller-financed deal often has a shape set by the Act, not just by cash flow.
Growing earnings during the hold multiplies through at whatever multiple the buyer paid on entry — that part is arithmetic, not a forecast. The harder part is that growth inside a small, owner-dependent business is rarely a smooth line: it depends on customer retention through a management transition, on capacity that was underused at the time of purchase, and on whether the diligence-stage growth story survives contact with the first year of new ownership. None of that is specific to any one province or sector; it is the reason a buyer’s underwriting case and a seller’s pitch deck are rarely the same document.
The multiple a buyer pays and the multiple a buyer eventually sells at are two separate numbers, set years apart, by different buyers with different information. Deavo’s own published sector median SDE multiples range from 2.1× for restaurants to 3.6× for manufacturing — “illustrative medians for research context only… not an appraisal” — and nothing in the Canadian data available anywhere states a market-wide multiple, a typical direction of travel, or a benchmark rate of expansion or compression over a multi-year hold. That is not a gap in this article; it is the honest state of the published record. A buyer who bakes multiple expansion into an underwriting model is, in effect, forecasting a number nobody publishes and nobody can.
A worked example (illustrative parameters, not a benchmark)
A target earns $600,000 of EBITDA and is bought at an entry multiple of 5.0× — a scenario parameter chosen to make the arithmetic legible, not a stated market rate — for an enterprise value of $3,000,000. The buyer funds it with $1,000,000 of equity and $2,000,000 of acquisition debt. Over a five-year hold, EBITDA grows to $780,000 (30%) and the debt is paid down to $500,000. At exit, the business is sold at 5.2× — a small, openly hypothetical uptick, not a forecast — for an exit enterprise value of $4,056,000, less the $500,000 of remaining debt, leaving $3,556,000 of exit equity against $1,000,000 put in: a 3.56× return on equity. Decomposed: debt paydown contributed $1,500,000 (the $2,000,000 to $500,000 reduction); EBITDA growth contributed $900,000 (the $180,000 EBITDA increase at the 5.0× entry multiple); and the multiple move contributed $156,000 (the 0.2× increase applied to the $780,000 exit EBITDA). The three add up to the full $2,556,000 gain — and two-thirds of it came from a lever the buyer actually controlled.
A buyer who cannot separate the three levers cannot tell a repeatable process from a market tailwind. The practical use of the decomposition is upstream, at the point a target is selected: a business whose story depends on multiple expansion is a bet on something nobody in this market can benchmark; a business whose story depends on debt paydown and earnings growth is a bet on financing terms and operating performance, both of which are visible before signing. See board composition after an investment for how governance rights get negotiated once that equity stake exists, and structuring investor returns on a single deal for how the legal documents actually allocate a gain like this one among more than one investor.
It removes one source of risk — there is less multiple compression to absorb if the exit market is worse than the entry market — but it says nothing about debt paydown or EBITDA growth on its own. A cheap entry multiple attached to a business with no realistic growth path and no financeable debt structure is not, by itself, a better outcome than a fuller price attached to a business that can actually delever and grow. See refinancing the acquisition debt two years later for how the debt side of that picture typically changes mid-hold.
Yes, and it is common in a difficult hold: EBITDA can grow while the exit multiple compresses, or debt can amortize on schedule while earnings stall. The decomposition does not promise the levers move together — it exists precisely so a buyer can see when they don’t, rather than reading a single blended number that nets the good news against the bad.
As a reference point for what an entry multiple might reasonably be in a given sector today — deavo’s six-sector medians are the only Canadian source that publishes anything at all here — not as a forecast of what the exit multiple will be years later. Those are two different questions, and nothing in the Canadian record answers the second one.
A short call maps your target’s realistic financing mix — CSBFP, bank debt and vendor financing — against the return it needs to produce.
Nobody publishes Canadian transaction data, so every valuation in this country quotes an American benchmark. We are building the Canadian one — multiples, asking-to-sale spreads and days on market, by sector and by city. Leave an email and you will see it first.
No pitch, no listings. One email when the first report lands.
