Most numbers that circulate about the Canadian small business market are not published anywhere. A short list of sources actually is — knowing which is which is the whole skill.
Key takeaways
STEP 01 OF 10
This is the single most load-bearing public dataset for this hub. As of December 2024, ISED reports 1.10 million employer businesses in Canada, of which 1.08 million (98.2%) were small businesses, 16,953 (1.5%) medium-sized, and 3,380 (0.3%) large, with a province-by-province breakdown down to individual counts — Ontario at 418,322 total employer businesses, Quebec at 233,235, British Columbia at 173,246, and so on through every province and territory. Use this table wherever a market-size figure is needed, rather than estimating one.
STEP 02 OF 10
The same ISED release reports that between 2017 and 2021, an average of 18,480 goods-producing and 62,550 services-producing small businesses were created annually, against 16,880 and 57,629 exits respectively, with survival to 21 years running 28.5% in goods-producing and 23.0% in services-producing sectors. This is real, dated, citable evidence for any claim about business turnover or succession pressure — use these figures rather than an unsourced estimate of how many owners are approaching retirement.
STEP 03 OF 10
CVCA's market report gives dated, sourced deal data: PE activity of $12.7 billion across 252 deals (deal count down 24% year over year, with four privatizations accounting for $7.25 billion, 57% of capital deployed), and VC activity of $2.69 billion across 250 deals (dollars up 17% year over year, deal count down 8.8%), plus $10.8 billion across 33 PE exits and 18 venture-backed exits totalling $716 million in disclosed value. Note the report was found via CVCA's own sitemap, not a linked page from the site's main navigation — the same discovery approach is worth trying on other institutional sites before concluding a figure is not published.
STEP 04 OF 10
Deavo publishes median SDE multiples by sector — 2.1× restaurants, 2.4× retail, 2.9× trades, 3.0× health & wellness, 3.2× professional services, 3.6× manufacturing — labelled “illustrative benchmark ranges—Canadian small-business transactions,” drawn from “Statistics Canada and ISED small-business statistics.” These are real, sourced figures, but the disclaimer is load-bearing: reproduce it with the figure every time, and never present a stated median as a range or a range as a median.
STEP 05 OF 10
Two of deavo’s own article titles promise figures they do not deliver: its trades-pricing and SME M&A outlook pieces name a topic in the title without stating a number in the body. Its CSBFP article deliberately avoids quoting current programme limits, and its lifetime capital gains exemption article confirms the exemption exists without stating a dollar amount. These are useful negative facts — knowing a source promises nothing is as valuable as knowing what a source actually states, because it stops a drafter from assuming a number exists somewhere on the page that a closer read does not support.
STEP 06 OF 10
Deavo’s listings pages show inventory counts by province and industry, but these describe what is listed on one platform, not the size of the underlying market — Quebec and Nunavut, for instance, show zero active listings on the platform despite being real, populous markets, which reflects platform adoption, not the absence of businesses for sale. Never cite a platform's own listing count as evidence of regional deal flow.
At a snapshot taken in August 2026, deavo’s own Ontario listings page showed roughly 57% of the platform's total inventory concentrated in that one province — a live, changing count rather than a fixed statistic, worth re-checking rather than repeating as a stable figure. It says more about where the platform has traction than about where Canadian business-sale activity is actually concentrated — ISED's business-count table in Step 1 is the more defensible source if the underlying claim is about market size rather than platform adoption.
STEP 07 OF 10
A StatCan “Daily” release renders server-side and is directly citable with its own release date. The interactive StatCan data tables behind many of those releases are client-rendered and are not reliably fetchable the same way — when citing StatCan, prefer the Daily release itself over a link to the underlying interactive table.
STEP 08 OF 10
A dated dataset stays accurate to its own release date, but combining two datasets from different years into one comparison can silently misstate a trend — confirm both figures share a comparable reference period before drawing a conclusion from the difference between them, and state the release date next to every figure rather than implying it is current.
This matters most for deal-count and deal-value figures, which move meaningfully year over year — CVCA's own H1 2026 report states PE deal count down 24% and VC deal count down 8.8% against the prior year, which means an H1 2025 figure quoted without its own date attached, next to an H1 2026 figure, would understate how much the market actually moved between the two periods.
STEP 09 OF 10
No Canadian source publishes a count of businesses currently for sale, a typical days-on-market figure, a standard PE carry or hurdle rate, an aggregate dry-powder figure, or a typical business-brokerage commission rate. Where a page needs one of these, state the mechanism instead of a number, or use an explicitly labelled hypothetical for arithmetic only — see choosing and briefing an intermediary for a worked example of exactly that distinction.
This absence is itself worth stating plainly rather than working around silently. A reader who has seen the same unsourced "2-and-20" or "typical multiple" figures repeated across marketing content benefits more from being told no Canadian body publishes one than from a drafter quietly declining to mention the gap at all.
STEP 10 OF 10
Where a figure is available both from the original regulator or industry body and from a secondary site repeating it, cite the original — ISED’s own page, CVCA’s own report, deavo’s own methodology page — rather than a summary that may have dropped the disclaimer, rounded the figure, or missed the release date. See preparing the marketing documents for a sale for how a normalized-earnings worked example applies exactly this discipline to a single business’s own numbers, not just market-wide statistics.
From ISED's own December 2024 table: Ontario has 418,322 employer businesses and Canada has 1,099,521 in total. Ontario's share = 418,322 ÷ 1,099,521 = 38.05%. British Columbia has 173,246, for a share of 173,246 ÷ 1,099,521 = 15.76%. Quebec has 233,235, for a share of 233,235 ÷ 1,099,521 = 21.21%. Those three provinces alone account for 38.05% + 15.76% + 21.21% = 75.02% of all Canadian employer businesses.
This kind of derived figure — a share, computed from the published table — is safe to state because it is arithmetic performed on a real, dated, sourced figure. What is not safe is inventing a similar-sounding figure with no table behind it, such as a province's share of "businesses currently for sale," which no source publishes.
ISED's business count and CVCA's deal figures describe the actual Canadian market, gathered through StatCan's Business Register and CVCA's own member and transaction reporting respectively — they are market-wide by design. Deavo's listing counts describe what sellers have chosen to list on one specific platform, which is a function of that platform's adoption, not the underlying market's size.
The distinction matters most in a province like Quebec, where ISED shows a large, real business population but deavo shows zero listings — the correct reading is that deavo has low adoption in Quebec, not that Quebec has no businesses worth buying or selling. Citing the wrong one of these two figures for a “market size” claim produces a conclusion the data does not actually support.
ISED's Key Small Business Statistics 2025 gives an exact, dated employer-business count by province as of December 2024 — the most current published breakdown available.
No — CVCA's own Canadianized ILPA templates state no fee rate, hurdle rate or carry benchmark anywhere. Describe the mechanism (a stated fee-base stepdown, a European or American waterfall, a catch-up) rather than citing a rate, unless it is an explicitly declared demonstration parameter.
No — deavo's own disclaimer states these are illustrative benchmark ranges for research context, not an appraisal. Use them as one data point alongside other valuation methods, not as a substitute for a professional valuation.
The table cited here is as of December 2024, published in the Key Small Business Statistics 2025 release — confirm no more recent edition has since superseded it before citing the same figures in new work.
A secondary summary can drop the release date, round a figure, or omit a caveat present in the original — citing CVCA's own report directly, as found through its sitemap, keeps the figure traceable to its source and dated correctly.
Not reliably — the interactive tables are client-rendered and are not consistently fetchable the same way a Daily release's server-rendered HTML is. Prefer the Daily release itself wherever one exists covering the figure needed.
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.
