The Canadian industry with the most to lose from not being named by assistants, and how to size the exposure
Why high value professional and regulated services carry the largest Canadian exposure, the three factors that set it, and an arithmetic you can run on your own numbers today.
Published by AI Knows Us · Updated 22 October 2026
High value Canadian professional and regulated services carry the most exposure: law, accounting, financial advice, specialist consulting, specialist medical and commercial insurance. Not because those buyers use assistants more, but because three factors multiply in those categories and almost nowhere else. The deal value is high, the shortlist is short, and the buyer often has no prior relationship to fall back on. Where all three hold, missing a three name shortlist removes you from the whole decision, and one lost engagement costs more than a year of the work it would have taken to be on it. Here is how to put a Canadian number on your own exposure rather than accept a ranking.
The three factors, and the arithmetic
| Factor | Why it multiplies the exposure | Your number |
|---|---|---|
| Deal value | One lost engagement in a high value category exceeds the whole cost of the remedy | Average first year value of a new client, in Canadian dollars |
| Shortlist length | A three name answer excludes everybody else, unlike a page of search results | How many suppliers your buyers actually evaluate |
| No prior relationship | A buyer with a trusted supplier does not ask. A buyer without one does | Share of new clients who arrived without a referral |
The arithmetic is deliberately simple, because a complicated model here would be false precision. Average first year client value, times the number of new clients a year who arrived without a referral, is the pool of revenue passing through a discovery process you do not control. Then ask what share of that discovery now includes an assistant. You do not need a national statistic; you need an estimate you would defend, and the sensitivity is what matters. If the exposure is material at 10 per cent, the work is justified without arguing about whether the true figure is 10 or 40.
Why this beats ecommerce as an exposure
It is counterintuitive, because consumer shopping is where assistant use is most discussed. A consumer who does not see a product in an assistant answer can still find it on a marketplace, in search results, on a shelf or through an ad, and the transaction is small and repeatable. A Canadian business buying a legal or accounting engagement runs a short, high stakes, infrequent process, often with a procurement step requiring nine documents. The shortlist is the decision, and removal from it is removal from the revenue with no second surface to catch it.
What the Canadian readings say about this category
On 22 October 2026 we put a 198 question Canadian buyer bank in 33 batches to the ChatGPT and Gemini product surfaces through a data provider, with Canada named inside the text of every question because the provider accepts gl, hl and location parameters and ignores all three. Batch 33 names us in the question and measures a ceiling, excluded from every figure. Sixty four blind readings, 32 per engine, 2,992 citation URLs across 974 distinct hosts.
Not one Canadian professional practice was named anywhere in that set. Thirty three of the 64 readings named no supplier at all. Software took 48 namings, led by Semrush at 16, HubSpot at 12 and Ahrefs at 9, and six Canadian marketing agencies were named 21 times. For a Canadian practice the exposure is therefore not that a rival is being named instead; it is that the category is answered in generalities, so the shortlist slot is unclaimed rather than lost.
That distinction changes the cost of the remedy sharply. Displacing three named incumbents is slow and expensive. Being the first specific, checkable, dated answer in a category where the engine currently names nobody is neither.
The constraint, and the assets available inside it
Canadian practices market under provincial rules that are real and that vary. Law society advertising rules restrict comparative and outcome claims and differ by province. Health professions are governed by provincial colleges. Insurance is licensed provincially with its own advertising requirements. The Competition Act's misleading representations provisions apply over all of it, administered by the Competition Bureau, amended in 2024, with a private right of access to the Competition Tribunal for certain deceptive marketing conduct from June 2025. CASL governs commercial electronic messages, and Quebec's Charter of the French Language reaches advertising and public communications in that province.
What those restrict is promises, comparisons and undisclosed endorsements. What remains fully available is a dated fee position in Canadian dollars with its tax treatment, a specification of what an engagement includes, and a counted figure about your own practice area with a denominator and a date. Those three are the assets the engines quoted elsewhere, and they are the ones most Canadian practices have not published.
How to size and start this quarter
- Run the arithmetic with your own three numbers and write the figure down. That is the business case and it takes ten minutes.
- Take a baseline. Twelve buying questions in your practice area with Canada or the province named in the text, both engines, saved with sources and dates, plus a French set if Quebec applies.
- Count how many of those readings name nobody. Those are the unclaimed slots and the cheapest work in the programme.
- Publish the fee position and the engagement specification, cleared by whoever already clears your advertising.
- Re-read at ninety days with the same questions and denominator, reported by engine.
If your practice is an accounting firm, the demand pattern follows the calendar and that is worked through in tax season leads and advisory clients are bought differently. If it is a law firm, the page structure that answers a client's situation rather than your service list is in pages by situation, written for someone in distress.
Questions about Canadian exposure to not being named
Is my Canadian industry really the most exposed?
Run the arithmetic rather than accept the ranking. Three numbers you already hold will tell you whether the exposure is material, and the ranking only says where to look first.
What if assistants name nobody in my Canadian category?
That is the better position, because the slot is unclaimed. In our reading 33 of 64 named no supplier, and taking an unclaimed slot is cheaper than displacing an incumbent.
Can a regulated Canadian practice publish enough to be named?
The rules restrict promises and comparisons rather than facts. A dated fee position, an engagement specification and a counted figure about your own practice area are all available and all quotable.
How urgent is this?
Treat it as a quarterly measurement. What makes it urgent in a specific practice is a short shortlist and a high deal value, which is the arithmetic rather than the calendar.
Do I need all five engines?
Start with two and say which two. Five surfaces are sold in this category, ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews, and reading two honestly is a real measurement.
Turning a Canadian exposure estimate into a measured position
An exposure figure is a business case; a reading is a position. On 22 October 2026 our Canadian readings named no professional practice in 64 readings and named nobody at all in 33 of them, which is worth knowing before a shortlist forms rather than after. AI Knows Us reads ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews with Canada named inside the question, reports each engine separately with its denominator, and keeps the raw answer text so progress can be audited. Outside the United States and the United Kingdom the cheapest plan that reads all five engines is Growth at 199 dollars a month with 100 buyer questions watched and 1,600 credits, read on aiknowsus.com/pricing and aiknowsus.com/llms.txt on 22 October 2026.