A T2 return was prepared to minimize tax, not to show a buyer what the business actually earns. Recasting it into seller's discretionary earnings is the step that turns a tax filing into a valuation input — and it is where sellers most often, deliberately or not, tilt the number.
Key takeaways
STEP 01 OF 10
Work from the corporation’s T2 return and the financial statements actually filed with it, not a spreadsheet the seller or their broker prepared for marketing purposes. A management summary can quietly present figures that were never filed, or that reflect a different fiscal period than the one under discussion — reconcile every recast figure back to the filed document before building on it.
Where the corporation keeps books that differ materially from what was filed for tax purposes, treat that gap itself as a diligence finding, not a formatting quirk — it is exactly the kind of pattern covered in uncovering tax exposure inside a target corporation.
STEP 02 OF 10
Deavo’s own methodology defines seller’s discretionary earnings as "profit plus the owner’s salary, perks, interest, depreciation and one-time costs" (deavo.ai/valuation). The key word is plus the full owner’s salary — not the excess over a market-rate replacement wage. SDE represents the total economic return available to a single working owner-operator, on the assumption that a buyer stepping into that role will draw the same kind of full return the business currently generates.
This is the point recast exercises most often get wrong by borrowing conventions from a different method: an EBITDA-style recast typically adds back only the excess of owner compensation over a market-rate replacement salary, because EBITDA assumes a hired, arm’s-length manager remains in the business. Mixing the two conventions in one recast produces a number that is neither a correct SDE nor a correct EBITDA figure.
STEP 03 OF 10
Work through net income before tax and add back, individually and with a documented reason: the owner’s full salary and bonus as reported; personal-use perks run through the business (a vehicle, insurance, travel); interest expense on business debt; amortization and capital cost allowance; and genuinely one-time, non-recurring costs such as a settled lawsuit or a one-off relocation.
Resist the temptation to bundle these into a single "adjustments" line. A buyer’s advisor will want to see each add-back individually, with the source document it came from — a bundled figure invites more scrutiny, not less, because it looks like something is being hidden inside it even when nothing is.
Keep a running total as you go, and reconcile it against the reported net income at every step — a recast that drifts from the source figures without anyone noticing where the drift started is much harder to defend later than one built line by line with a checkpoint after each addition.
STEP 04 OF 10
Recasting runs in both directions. A one-time government subsidy, an insurance settlement, or a gain on the sale of an unrelated asset can inflate a single year’s net income in a way that will not repeat under new ownership — those items should be subtracted from the recast, not left in as though they were ordinary operating profit.
A recast that only ever adds back and never subtracts should be treated with suspicion by whoever is reviewing it — a business rarely has one-time costs but zero one-time gains across several years of filed returns.
STEP 05 OF 10
The test for any add-back is simple to state and easy to get wrong in practice: would this specific cost or benefit disappear, or would it continue, if a new owner stepped in tomorrow? A personal vehicle lease disappears. A below-market family-member salary that would need to be replaced with a market-rate hire does not disappear — it changes form, and adding it back without accounting for that replacement cost overstates the business’s real discretionary earnings.
A useful cross-check from the tax side: a bad-debt write-off is only a legitimate deduction under ITA s. 20(1)(p) where the debt was "established by the taxpayer to have become bad" and "included in computing the taxpayer’s income" in that year or an earlier one. A recast that adds back a bad-debt expense the corporation never actually claimed correctly under that rule is adding back something that was not a legitimate deduction to begin with — check the underlying tax treatment, not just the bookkeeping entry.
STEP 06 OF 10
Take a target corporation reporting net income before tax of $140,000 on its filed T2. Add back the owner’s reported salary and bonus of $180,000; a $14,000 vehicle lease used primarily for personal purposes; $22,000 of interest expense on business debt; $48,000 of amortization and capital cost allowance; and $35,000 of legal costs tied to a one-time lawsuit settled that year. Subtract an $18,000 one-time government wage subsidy that will not recur.
The recast SDE is $140,000 + $180,000 + $14,000 + $22,000 + $48,000 + $35,000 − $18,000 = $421,000. That is more than three times the reported net income — not because the business was hiding profit, but because a filed corporate tax return is built to minimize taxable income, not to show a buyer the full economic return available to an owner-operator.
STEP 07 OF 10
At a trades-and-services median multiple of 2.9× SDE, the $421,000 figure from Step 6 produces a valuation midpoint of roughly $1,220,900 — the full method, including the multiple table and the cross-check step, is covered in valuing a Canadian owner-operated business. Deavo’s own capital-stack tooling switches from an SDE basis to an EBITDA basis for underwriting purposes at roughly $1 million — "lenders underwrite micro deals on SDE and larger deals on EBITDA" (deavo.ai/financing).
A recast landing close to that switch point is worth a second look: confirm which convention a lender or buyer will actually apply before assuming your SDE-based add-backs carry through unchanged into an EBITDA-based underwriting review.
STEP 08 OF 10
A recast built from numbers that cannot be reconciled to what was actually filed is a different problem entirely from an aggressive but honest add-back schedule. Real Canadian acquisition diligence has surfaced targets carrying genuinely different internal figures than what was reported to the CRA — a pattern that is not a recast dispute, it is a diligence finding that should stop the process until it is fully understood.
A related pattern worth watching for specifically: a rebate or supplier-credit arrangement that flows through the income statement in a way that inflates reported gross margin without a corresponding, sustainable improvement in the underlying business. Treat any recast you cannot trace, line by line, back to the corporation’s own filed return and supporting documents as unverified — regardless of how professionally it is formatted or how confidently it is presented.
STEP 09 OF 10
Every add-back and subtraction should carry a one-line explanation and a reference to its source document — the specific invoice, the specific payroll record, the specific settlement agreement. This is not just diligence hygiene; a well-documented recast is also the schedule a buyer’s lender will want to see before relying on the SDE figure for financing purposes.
A recast that cannot survive a line-by-line challenge from a skeptical reader is not ready to be used as a valuation input, however defensible the underlying logic feels to the person who built it.
STEP 10 OF 10
Where a seller or a broker presents an already-recast SDE figure, rebuild it independently from the filed return rather than accepting the summary number. Confirm every add-back individually against Step 5’s test — would this genuinely disappear under new ownership — and confirm every subtraction was actually applied, not just claimed.
A recast built by the party selling the business carries an obvious incentive to lean toward inclusion rather than exclusion on any judgment call. An independently rebuilt figure that lands close to the seller’s number is a good sign; one that diverges materially is exactly the kind of finding that belongs in the negotiation, not glossed over.
The distinction in Step 2 is worth restating on its own, because conflating the two methods is the single most common recast error this guide is written to prevent. SDE assumes one owner-operator drawing the business’s full discretionary return; EBITDA assumes a professionally managed business where an owner’s compensation is replaced at a market rate rather than added back in full. A recast that adds back a full six-figure owner salary and then applies an EBITDA-style multiple to the result — rather than an SDE-style multiple — will overstate the valuation, sometimes substantially.
Confirm which method a specific multiple or benchmark was built for before applying it to your recast figure — the two are not interchangeable inputs to the same formula.
Two patterns recur often enough in Canadian small-business acquisitions that they are worth naming specifically. One is a target whose internally tracked figures diverge from what was actually reported to the CRA — effectively two different sets of numbers describing the same business, only one of which was filed. The other is a rebate or credit arrangement that flows through as revenue or margin in a way that flatters the recast without reflecting a repeatable improvement in the underlying operation.
Neither pattern announces itself in a management summary. Both surface only when a recast is rebuilt independently from source documents rather than accepted as presented — exactly the discipline Steps 9 and 10 of this guide describe. See three years of statements, two sets of books and the rebate cheque that inflated gross margin for how each pattern actually played out in a real deal.
No. SDE adds back the owner's full compensation and assumes one owner-operator; adjusted EBITDA typically adds back only the excess over a market-rate replacement salary and assumes a professionally managed business. Applying an EBITDA multiple to an SDE figure, or vice versa, produces a distorted valuation.
Treat the add-back as unverified and exclude it from the recast until documentation is provided. An add-back a seller cannot support with a source document should not carry the same weight as one that is fully traceable.
Use it as a starting point, not a finished answer. Independently rebuild the figure from the corporation's actual filed tax return, since the party selling the business has an obvious incentive to lean toward inclusion on any judgment call.
There is no statutory cutoff, but market practice commonly shifts toward an EBITDA-based approach as earnings approach roughly $1 million, which is also where lender underwriting conventions tend to switch basis.
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 as each measure is published.
