Anonymised, illustrative composite. No document leaked. The teaser itself was specific enough to identify the business to anyone who could count.
At a glance
A boutique sell-side advisor prepared a blind teaser for a specialty manufacturer on Prince Edward Island ahead of a planned auction process. Blind teasers are meant to generate interest without disclosing who the business is, so prospective buyers can decide whether to sign a non-disclosure agreement before learning the seller’s identity. This one circulated for eleven days before the seller’s largest customer called to ask, directly, whether the business was for sale.
The teaser was accurate and, on its own terms, well written: a precise trailing-twelve-month revenue figure, a narrow sector description specific enough to distinguish the business from adjacent categories, and the province. Individually, none of those three facts identifies a business. Together, in a jurisdiction with few businesses to begin with, they very nearly do.
The advisor had used the same teaser template on three prior mandates, all in Ontario, without incident — a precise revenue figure and a narrow sector line had never been enough, on their own, to identify a seller among Ontario’s 418,322 businesses. Nobody on the deal team re-examined the template’s assumptions when the mandate moved to a market roughly eighty times smaller.
Prince Edward Island had 5,339 total employer businesses as of December 2024, per ISED’s Key Small Business Statistics — the smallest business count of any province by a wide margin, against Ontario’s 418,322 over the same period. It is worth being precise about which cut does the narrowing, because the obvious answer is the wrong one. ISED’s industry table counts 256,942 goods-producing employer businesses out of 1,099,521 nationally, so the goods-versus-services split alone removes only about three-quarters of the field — not almost all of it. Manufacturing specifically accounts for 47,237, roughly 4.3%. It is the narrowing to manufacturing, not the broad sector split, that strips out some 96% of candidates. ISED publishes those industry counts nationally rather than province by province, so applying that share to Prince Edward Island yields an order of magnitude — a couple of hundred manufacturers — and not a published local count. Layer in the teaser’s precise revenue figure — not a range, an exact number — against a provincial business base that small, and anyone in the industry with even rough knowledge of local competitors has, in practice, one candidate left to guess.
There is no statute at issue here — the rule is arithmetic, not law. A blind teaser’s purpose is defeated the moment its combination of facts narrows the candidate pool to a small enough number that informed industry participants can identify the seller without ever seeing a name. The smaller the denominator — a small province, a narrow sector, a specific size band — the fewer identifying data points a teaser can safely combine before it stops being blind.
The seller’s customer relationship survived the call, but only because the advisor got ahead of it — confirming, briefly, that a process was underway and reassuring the customer that service continuity was a term the seller intended to protect in any transaction. The advisor withdrew the original teaser and reissued a revised version the following week: revenue restated as a band rather than a precise figure, and the sector description broadened one level up the classification. The process continued with three of the original nine parties who had already made informal inquiries, none of whom the seller believed had been the source of the customer’s call.
The advisor also changed its internal process for any future small-province mandate: before a teaser goes out, run the same identifiability check the seller had to run only after a customer called — provincial business count, narrowed by sector and size band, checked against how many candidates remain. Where the remaining field is thin, widen the sector description or the revenue band before release, not after.
Once a process is underway, the harder version of this problem is a genuine competitor gaining access to material the teaser was never meant to precede — see a data room leak that reached a competitor. For a screening failure that killed a deal for the opposite reason — too much specificity in the buyer’s own financing math — see walking away in week two on a single number.
Had the customer relationship not survived the disclosure, the cost would not have shown up in the sale price directly — it would have shown up as a customer who began quietly qualifying a second supplier the moment ownership uncertainty entered the picture, eroding the very revenue base the sale process was trying to sell. In a small province, that kind of reputational cost is also harder to contain: word travels through a business community with 5,339 total members far faster than through one with 418,322.
The tell is treating “blind” as a property of the document rather than a property of the market it is released into. A teaser with the same three data points would have been genuinely anonymous in Ontario and was nearly a name-and-address in Prince Edward Island. Before finalizing a teaser, run the same test the buyer should have run first: how many businesses in this province, this sector, this size band, could this plausibly be?
A 30-minute call is enough to tell you whether AI pays for itself here.
