Anonymised, illustrative composite. A manufacturer being carved out of a family group paid a flat 4% of revenue for finance, IT and HR. The seller wanted the whole fee added back to EBITDA. The buyer built the real cost from the ground up instead.
At a glance
The target was a mid-sized Ontario manufacturer, one of three operating companies under a single family holding company. For as long as anyone could remember, each operating company had paid the holding company a management fee of 4% of its own revenue, and the holding company used that pool to run one finance department, one IT department and one HR department for all three businesses. Nobody had set the 4% by reference to anything — it was the number the family's original accountant had picked at incorporation, and it had never been revisited.
The buyer was acquiring only the manufacturer, as an asset purchase. Finance, IT and HR would not come with it. The seller's information memorandum treated this cleanly: it showed reported EBITDA, added back the full $720,000 management fee as a related-party cost that “will not recur post-closing”, and presented the resulting adjusted EBITDA as the number to put a multiple on.
A management-fee add-back assumes the buyer will simply stop paying the fee and incur nothing in its place. That is never true for finance, IT and HR — someone still has to run payroll, close the books, keep the network up and administer benefits on day one after closing. The real question is not whether the fee disappears. It does. The question is what it costs to replace what it was buying.
Nobody inside the group could answer that. The target's own controller reported into the holding company's CFO and had never priced an outside accounting function. The holding company's IT staff had never obtained a managed-services quote for the manufacturer alone. The 4% figure was an internal cash-allocation convention, not a benchmark against any external market.
Target revenue $18,000,000; the management fee was $720,000 a year (4% of revenue). The seller's model added the full $720,000 back to EBITDA.
The buyer's diligence team priced a standalone replacement function by obtaining real quotes rather than estimating: a controller plus two staff and a mid-market ERP licence for finance ($480,000), a managed-services-provider contract actually quoted for the target's server and user count for IT ($340,000), and a fractional HR lead plus an outsourced payroll bureau for HR ($230,000) — $1,050,000 in total, 46% above the fee it was replacing.
The gap between the fee added back and the cost actually required is $330,000 a year. At the deal's agreed 6.0x EBITDA multiple, that gap is worth $1,980,000 of purchase price if it is left unadjusted.
Two statutes are worth reading here precisely because neither one ends up governing, and both were checked rather than assumed away. The Income Tax Act's related-party pricing rule, s.69(1)(a), provides that “where a taxpayer has acquired anything from a person with whom the taxpayer was not dealing at arm's length at an amount in excess of the fair market value thereof at the time the taxpayer so acquired it, the taxpayer shall be deemed to have acquired it at that fair market value.” That rule needs an ascertainable fair market value for the thing acquired. A related-party lease can be tested against comparable market rent for similar space in the same area — an observable, quotable figure. A bundled, bespoke package of finance, IT and HR run for three affiliated companies at once has no comparable market price to test it against; nobody sells exactly that bundle to arm's-length strangers. Even taken at face value, s.69(1) supplies no workable benchmark for a service this specific.
The Excise Tax Act's own non-arm's-length pricing rule for GST/HST, s.155(1), looked more promising at first: it deems consideration for a non-arm's-length supply of “property or a service” to equal fair market value where the price charged is below it — which is exactly this fact pattern, a related-party service priced under what it costs to provide. But s.155(1) applies only where the recipient “is not a registrant... acquiring the property or service for consumption, use or supply exclusively in the course of commercial activities of the recipient.” The target is an ordinary manufacturer using finance, IT and HR exclusively in its own taxable operations — precisely the registrant s.155(1) excludes from its own reach. The provision exists, was on point on its face, and disapplies itself on these facts.
With both statutes checked and neither reaching the question, the correction is a modelling exercise: build the standalone cost bottom-up from real quotes, compare it to the fee being added back, and adjust normalized EBITDA by the difference — not by the fee.
The parties did not simply reject the seller's add-back; they replaced it with the built number. Normalized EBITDA was reduced by $330,000 a year relative to the seller's model, cutting the purchase price by $1,980,000 at the agreed multiple.
The agreement also added a twelve-month transition services agreement, priced at the group's actual cost of providing finance, IT and HR plus a 5% markup, giving the buyer runway to stand up its own functions rather than a hard cutover at closing. The TSA price was set from the same bottom-up figures used in the adjustment, not from the old 4% convention.
Accepting the seller's add-back at face value would have overpaid by $1,980,000 relative to the business's real standalone earning power, and the buyer would have discovered the gap the hard way — in the first year after closing, once the TSA (if any had been negotiated) ran out and actual invoices for finance, IT and HR came in 46% above the number the price had been built on.
The tell was that nobody could produce a quote. A defensible add-back can point to a market rate — a comparable lease, a market salary survey, a competing supplier's price sheet. Here, when the buyer asked the target's own controller and the holding company's CFO separately what finance, IT and HR would cost run alone, neither could answer, because the 4% figure had never been priced against anything external. A number nobody inside the seller's own organisation can defend is not a benchmark; it is a placeholder.
A 30-minute call is enough to tell you whether AI pays for itself here.
Canadian small-business transaction data is not published anywhere, so most valuations in this country quote an American benchmark. The Deavo–Treadstone Acquisition Index is a daily record of Canadian listings built to replace that: asking-price distributions by province and city are published now, and days on market, departure rates and asking-to-sale spreads follow as the series lengthens. Leave an email and we will tell you as each measure lands.
No pitch, no listings. One email when the first report lands.
