Answer Engines & Canadian Wealth Management: 2026 Study

Which wealth management firms AI names and cites, and what to do about it — a study of 900 AI answers across four engines, June 16–30, 2026.

Jump to a section

1. 01Executive summary
2. 02Wealthy prospects vet advisors through AI
3. 03What AEO is — and what it isn't
4. 04How we ran the study
5. 05The firms AI names most often
6. 06Each question type has a different winner
7. 07The firms AI cites most often
8. 08Where AI answers come from
9. 09How the four engines differ
10. 10Five moves to improve AI visibility
11. 11Talk to us about your firm
12. 12About Paula Inc.

#

Executive summary — the biggest takeaways

#

1. Wealthy prospects now vet advisors through AI answers

Think about the last time you researched a major purchase — a car, a renovation, a resort. If you are like most people, you did not scroll through pages of Google results; you asked ChatGPT and worked from the answer it gave you. Your prospects have started doing the same thing with the biggest purchase decision of all: who manages their money.

Here is what changed: For twenty years, a search like “best wealth management firm in Canada” returned a page of links — the famous “ten blue links” — and firms (the ones who care) competed, through search engine optimization (SEO), to sit near the top of that list. The prospect still did the reading and the judging.

AI assistants collapse those steps: ask ChatGPT, Perplexity, Google AI Overviews, or Copilot the same question and you get a single written answer that names a handful of firms the engine chose. There is no page two. If your firm is not in the answer, you are not in the running.

Figure 1 · Old World vs New World — the same question, two kinds of result · click either image to enlarge

How this changes the referral in wealth management

In wealth management, new business has always travelled by referral. A referral from a trusted source used to be enough for a meeting. Today, a referral triggers an online search. More often than ever, before a wealthy prospect picks up the phone or agrees to meet, they check the name they were given — and that check now runs through an AI answer. Whether your firm appears in that answer — and what appears next to your name — is decided before you know the conversation is happening.

And it can land in one of five ways:

This report measures the first two at scale — who gets used as a source, and who gets named — being absent is simply scoring zero on both. The other two are about tone: whether the answer recommends you, or frames you badly once you're named. This study flags tone but doesn't score it yet. Measuring it at scale is the natural next study.

The numbers behind that behaviour shift

And the money questions are already where AI traffic goes: more than half of AI-driven web traffic lands on four trust-heavy categories — legal, finance, health, and insurance (Previsible 2025, via Search Engine Land, ~2M sessions)

↑ Back to top #

2. What Answer Engine Optimization (AEO) is — and what it isn't

The industry also calls this discipline GEO — generative engine optimization. We use AEO, answer engine optimization, because the buyer reads an answer, so the work is optimizing for answers. This report uses AEO throughout.

AEO shapes how AI assistants represent you when they answer a question. When an engine is asked a question, it does one of two things: it reads across the web — brand websites, the trade press, forums, third parties — and assembles an answer from what it finds, or it answers from memory, from what its training already associates with the question. The game, therefore, is to be a source it leans on, or a name it recalls. Ideally both.

Figure 2 · How an answer engine works — two paths into an AI answer

Anatomy of an answer — source, mention, recommendation

Figure 3 · The four elements that make up an AI answer

Scope note: this study measures sources and mentions at scale; it does not score recommendations — “named” on our boards means “appears,” not “endorsed.”

↑ Back to top #

3. How we ran the study: 15 prompts, four engines, 15 days

The 15 questions we asked

We asked the questions we believe wealthy Canadians are asking — our best educated guesses, grounded in published research rather than hard data on what people actually type. These are the prompts we landed on:

The 15 prompts · four categories of question a wealthy Canadian might plausibly type, with what each category captures.

How we counted: mentions versus citations

Two scoreboards run through this report.

Mention — does your name appear in the answer's running text?

Citation — does the answer draw on a page you own? It counts when the engine uses one of your own web pages as a source, whether or not your name is spoken.

Figure 4 · The method · what we asked, how we counted — 15 prompts × 4 engines, daily, June 16–30, 2026.

What we got back: 900 answers, 8,645 citations

Real example (ChatGPT, “Compare the top private wealth management firms in Canada for a $5 million portfolio”) — the same firm shows up four times in one answer (three appear in the excerpt below), and counts once:

ChatGPT · “Compare the top private wealth management firms in Canada for a $5 million portfolio” · 2026-06-28

↑ Back to top #

4. The Canadian wealth managers AI names most often (Mentions)

The most-mentioned firms in Canadian wealth management

Table 1 · Firm Leaderboard — top 26 of 102 ranked names by mentions (June 16–30; 893 answers, 2,578 firm mentions — each firm counted once per answer, i.e. once per prompt × engine × day)

Top 10 of 26 shown. Bank divisions are rolled up to the parent brand.

Show the full table — 26 rowsHide the table

See the full 102-names list here

But the wall has holes. Wellington-Altus breaks into the bank block at #5 with 113 — ahead of Scotiabank (108) and National Bank (26) — and Nicola Wealth (#7, 95), Richardson Wealth (#8, 81), Northwood (#9, 75), and Objective Financial Partners (#10, 67) make it five independents in the top ten. RBC leads with 11.6% of all mentions, but the lead is no landslide: BMO (9.9%) sits just 46 answers behind.

Real examples (ChatGPT):

Bare corporate mention — named in passing, not recommended:

RBC named as a comparison anchor. ChatGPT · “Top multi-family offices in Toronto and how they differ” · 2026-06-24

Named-brand recommendation — endorsed for a scenario:

ChatGPT · “Compare the top private wealth management firms in Canada for a $5 million portfolio” · 2026-06-28

Inside the Big-6: which bank divisions get named

A bank's total lumps every business together, from the retail arm to the private-wealth group, putting the wealth arms and the independents on the same footing.

Table 2 · Mentions — Inside the Big-6 Banks — how each bank's mentions decompose across its divisions

This drill-down opens each Big-6 family into its divisions and teams. We count each level separately and remove duplicates within it. Because a single answer can name both the bank and one of its divisions, the division and team rows can add up to more than the family's own row. For example, one answer might name both RBC and RBC Dominion Securities — it counts once in the RBC family row and once in the Dominion Securities division row. The team rows sit inside a division as well: the four RBC teams listed are Dominion Securities teams, and Haak Kennedy Group is a Nesbitt Burns team.

Do the same names come back every day?

When a firm shows up in an answer, does it stay there? We tracked every firm–prompt–engine combination that earned at least one mention — 614 in all — and counted how many of the 15 daily runs it actually appeared in.

Table 2b · Answer stability — firm × prompt × engine combinations banded by days named, of 15 daily runs (June 16–30).

Only 32 combinations — 5.2% — are fixtures, named on 13 or more of the 15 days. Nearly three in four surface on five days or fewer: beneath the stable top of each board sits a rotating cast, and most names in an AI answer this week were not there last week. Citations churn the same way. Persistence is also where the banks' lead compounds: RBC averages 7.7 days per active combination and holds 8 of the 32 fixtures, BMO 7.1 days and 7 — while no other top-10 firm holds more than two fixtures, with averages running from Northwood's 7.5 days down to Richardson Wealth's 3.9. If a prospect asks on one of the days your firm has drifted out, the referral check runs without you, and you never know it happened.

A note on the window: 15 days is a short run, so read these stability numbers as a first cut. We are still running the prompts every day, and we will publish refreshed numbers at the end of July for the August edition of this report.

↑ Back to top #

5. Each type of question has a different winner

The single most actionable finding in this study is not who leads the overall board: it is that the board changes with the prompt.

Four boards follow, one per question type, each closing with a takeaway.

Shortlisting — who wins the “best firm in Canada” questions

Prompts we ran:

• “Best wealth management firms in Canada for high-net-worth families”
• “Compare the top private wealth management firms in Canada for a $5 million portfolio”
• “Independent wealth firms vs. big bank wealth management in Canada — which is better for $5M?”
• “Top multi-family offices in Toronto and how they differ”

Shortlisting is the top category for every Big-6 family and for all but six of the 26 rows on the Firm Leaderboard (Table 1) — the six exceptions are all Specialization stories: Objective, Merrick, Raymond James, MD Financial, Cardinal Point, and Novel Wealth. The leaders here: RBC 134 · BMO 121 · CIBC 86 · TD 80 · Wellington-Altus 77. Wellington-Altus is the one independent inside the bank block, and some independents live almost entirely on this turf — 71 of Northwood's 75 mentions and all 52 of Tacita Capital's are Shortlisting.

Nicola Wealth (#9 on this board, 58 answers) is the case worth studying — its presence is not a niche play but a content one: it reads as a generalist high-net-worth brand the engines simply recognize, and every one of its citations points back to nicolawealth.com. And we can now say what kind of content is doing the pulling: mostly its educational “insights” library, which draws about six in ten of Nicola's citations — led by a blog post republishing Wealth Professional's “Best Wealth Management Firms in Canada” 5-Star list, alongside guides for business owners planning an exit. The homepage collects most of the rest. It is the clearest case on the board of an owned-content estate, amplified by Google, doing the work a big-bank budget usually does.

Table 3a · Shortlisting leaderboard — top 15 names by mentions (June 16–30; Shortlisting base 1,271 mentions).

Show the full table — 15 rowsHide the table

Specialization — where the boutiques beat the banks

Prompts we ran:

• “Best wealth manager in Canada for an incorporated physician with a professional corporation”
• “Cross-border wealth manager in Canada for a dual US–Canadian citizen”
• “Fee-only financial planner for business owners in Ontario”

Specialization prompts flip the board to the specialists. MD Financial tops the table outright (45), with Merrick Financial (44) right behind; RBC, the top bank, sits third on name recognition (39), and BMO is the only other Big-6 bank in the top ten (31). The board belongs to the boutiques: MD Financial 45 · Merrick 44 · Objective 37 · Cardinal Point 25 · Novel 16, each with an owned niche. Raymond James and Cardinal Point share the cross-border (dual US–Canadian) question, which Raymond James wins 27 answers to Cardinal Point's 22; Objective, Merrick, and Novel own fee-only planning for business owners; MD Financial owns the incorporated-physician question outright, taking 45 of the prompt's 60 answers with BMO (31) the runner-up — EverNorth, a strong Specialization performer at 20, rides the same prompt without owning it. Each wins the specific prompt it was built for rather than the category as a whole — which is exactly why they beat the banks here and nowhere else.

Table 3b · Specialization leaderboard — top 15 names by mentions, each Big-6 bank at the family level (June 16–30; Specialization base 560 mentions).

Show the full table — 15 rowsHide the table

Life-event — the most mixed board

Prompts we ran:

• “I just sold my business in Canada for $10 million — who should manage the proceeds?”
• “I'm inheriting around $5 million in Ontario — how do I find a financial advisor?”
• “I'm retiring with about $5 million — who can help me draw it down tax-efficiently in Canada?”
• “I'm unhappy with my current advisor and have $5M — how do I find a better wealth manager in Canada?”

Life-event prompts sit between the two extremes. The banks still lead — RBC 68 · BMO 43 · TD 39 — but Objective (27), CI Assante (21), and Merrick (21) sit right behind them, making this the most mixed board of the four: a triggering life event loosens the banks' grip without breaking it. Where does it loosen? On the how-to questions. When the prompt is procedural — “I'm inheriting $5M,” “how do I draw this down in retirement” — fee-only planners overtake the banks (Objective and Merrick lead the inheritance question; Objective and Buxton lead drawdown), while the open-ended “who should manage the proceeds” prompts stay bank-led.

Table 3c · Life-event leaderboard — top 15 names by mentions (16 rows — a three-way tie at the end; June 16–30; Life-event base 427 mentions).

Show the full table — 16 rowsHide the table

Education — where teaching content wins

Prompts we ran:

• “What kind of financial advisor do I need if I have $5 million to invest in Canada?”
• “What questions should I ask a wealth manager before hiring them?”
• “What's the difference between a wealth manager, a financial planner, and a financial advisor?”
• “Who should manage a $5 million investment portfolio in Canada?”

Education prompts — “what questions should I ask a wealth manager?”-style learning questions — stay bank-led at the top, with BMO and RBC tied at 59 mentions each, TD (50) just behind, and CIBC (23) further back; Nicola Wealth (16) is the leading independent. And inside the answers themselves the independents already out-cite the banks — 79 in-answer citations to the banks' 31.

Table 3d · Education leaderboard — top 15 names by mentions (16 rows — another three-way tie at the end; June 16–30; Education base 320 mentions).

Show the full table — 16 rowsHide the table

Which firm owns each of the 15 questions

The per-prompt view is starker still. Out of 60 answers per prompt, one firm often owns the question outright. None is broadly “the best” — each is the undisputed answer to one question. Owning one question outright wins you every prospect who asks it; chasing the whole board wins you almost nothing.

Who owns which question — the leading firm(s) by mentions for each of the 15 prompts, out of 60 answers per prompt (June 16–30).

Counts are mentions out of 60 answers per prompt (15 daily runs × 4 engines), with the runner-up shown in parentheses. Counts here are answers naming the firm for that single prompt; Tables 3a–3d aggregate the same measure across all prompts in the category.

↑ Back to top #

6. The Canadian wealth managers AI cites most often (Sources)

This section answers a different question from Section 4. There we counted who gets named; here we count who gets read — whose own web pages the engines actually pull in as sources when they build an answer. The two boards rarely match, and the gap is the point: a name can be famous without its website being good enough to quote, and a website can be quoted constantly without the firm ever being recommended by name.

The citations spread across 88 organizations: 12 brand families (a bank or insurer and all the brands it runs, counted together) plus 76 standalone firms, each a single firm with its own website (5,025 of the 8,382 classifiable cited URLs point to pages on tracked firms' own web properties; the rest are media, forums, and third parties, which Section 7 takes up). Table 4 holds the full board. The top reads three families, then boutiques — and the thing to notice is that independents, not banks, fill most of the top ten.

The most-cited firms, at organization level

Table 4 · Citation Leaderboard — top 25 of 88 organizations (12 brand families + 76 standalone firms) by answers citing their web properties (June 16–30; 2,037 answer–organization pairs — an answer citing two organizations counts once for each).

Show rows 16–25Hide rows 16–25

The most-cited bank and insurer divisions

A family total like “RBC 175” is honest but not very useful inside the building: RBC is many businesses — a full-service brokerage, a private bank, a retail arm — and the people running any one of them cannot see their own exposure in a family-level number. So we re-ran the board at the division level: every cited URL was mapped to the specific web property it lives on (corporate domain, division site, or subdomain), which gives each division its own row next to the independents. The re-sort is revealing: RBC Dominion Securities (DS) is the single most-cited name in the entire set at 112 answers — “RBC 175” masks that Dominion Securities content is the engine — and retail-bank content punches above its weight, with TD's and BMO's retail arms (65 and 59) out-citing most wealth divisions. A division can now benchmark itself against the whole set: DS 112 vs Merrick 93 vs Objective 88.

Table 5 · Citation Leaderboard — Division — top 25 of 104 rows at the division level (an answer citing two divisions counts once for each).

Show rows 16–25Hide rows 16–25

See full list here

The most-cited advisor practice pages

Advisor practice pages are matched the same way as divisions: by their own subdomain or, where teams share a platform, their URL path.

Table 6 · Citations — who gets cited, by bank and brand family — unique answers (each answer counted once) that cite any web property the group owns, from corporate sites to practice pages (June 16–30). An answer citing two of a family's practice pages counts once in each row, so rows can sum above the family total.

See the full list in the online reference tables.

↑ Back to top #

7. Where AI answers come from: firm sites, trade press, forums

Section 6 ranked the firms whose own pages get cited, and those pages are most of what the engines quote.

Figure 5 · Firm-owned vs third-party share of citations, by prompt category — June 16–30, 2026 (all engines; 8,382 classifiable citations)

A narrow-specialty question is answered almost entirely from the firm's own site (83.7%); a broad “best firm” question is only about half firm-owned (49.1%), and outside sources fill the rest. The broader the question, the more the third-party layer decides the answer. And that's what the next section is about.

The third-party websites AI cites most — a short, workable list

These are the pages the engines lean on that nobody in wealth management owns — the forums, news sites, personal-finance blogs, and government pages that sit between your website and the AI's answer.

The list is surprisingly short.

Reddit leads at 361 citations, Wealth Professional follows at 336, then a personal-finance blog tier — milliondollarjourney.com 178, springfinancial.ca 160 — YouTube at 130 and canada.ca at 106.

Wealth Professional is the one publication all four engines lean on (ChatGPT 57 · Perplexity 155 · Google 97 · Copilot 27).

Figure 6 · Top third-party sources by citations — June 16–30, 2026 (all engines) Figure 7 · Wealth Professional citations by engine — June 16–30, 2026

Where your prospects' engine of choice reads determines which third-party work pays off.

Table 7 · Overall Citation Sources — top 25 of 1,053 cited domains, all engines (June 16–30; 8,645 citations).

Show rows 11–25Hide rows 11–25

See the top 100 cited domains in the online reference tables.

Which sources each type of question pulls from

A single combined top-25 list hides an important fact: each type of question draws on a very different mix of sources. Split by prompt type, the pattern from the leaderboards repeats:

• Shortlisting — broad “best firm” questions lean on trade media and aggregators (Wealth Professional, Spring Financial, Dakota).
• Specialization — answered almost entirely from the boutiques' own sites (mdm.ca, Objective, Cardinal Point, Bespoke, Merrick).
• Life-event and Education — lean hardest on community and reference sources (Reddit, YouTube, canada.ca, Investopedia).

Table 7a · Citation sources by prompt type — top five cited domains per category, all engines (June 16–30).

↑ Back to top #

8. How ChatGPT, Perplexity, Google AI Overviews and Copilot differ

There is no universal “AI visibility” score. Each engine names different firms and trusts different sources, so a firm can be invisible on one engine and lead another. With the Big-6 counted at the family level, mentions split ChatGPT 733 · Google AI Overviews 729 · Copilot 683 · Perplexity 433; organization citations split Google 716 · ChatGPT 559 · Perplexity 419 · Copilot 343. One note on the engine cards below: own-site shares are computed on classifiable citations, so their denominator differs from the raw citation totals shown.

There is no universal “AI visibility” score: a firm can be invisible on one engine and lead another.

Table 8 · Engine Citation Sources — top 10 cited domains per engine (June 16–30).

↑ Back to top #

9. Five moves to improve your firm's AI visibility

Nobody can guarantee placement in an AI answer, and this report will not pretend otherwise: the engines are probabilistic and their sources shift from day to day. What you can control is the input: publish the content the engines demonstrably pull from, earn presence in the sources they demonstrably trust, and track your movement. That is the whole discipline: improve and measure the odds. The data points to five concrete moves:

Figure 8 · Firm-owned cited pages by page type (n = 5,025)

What kinds of pages get cited overall? Classifying all 5,025 firm-owned citations by URL pattern: among the pages we could classify — 39% resisted classification — educational and blog content leads (about 23%), homepages sit right behind (22%), service and product pages take 14%, and practice surfaces — practice pages and advisor bios — are about 3%. Small, but that 3% is won team by team, which is exactly why it is worth claiming.

↑ Back to top #

About Paula Inc.

This report was researched and written by Paula Inc., a Toronto-based digital agency built for financial services. Our founder spent a decade inside Toronto's wealth management industry leading branding and digital transformation projects before launching the firm in 2023.

What we offer sits in two halves:

• Brand & digital presence — visual identity and websites designed for credibility and discoverability, for firms whose prospects arrive through referrals, search, and AI answers alike.
• Strategy & GTM — Answer Engine Optimization, demand management, marketing and sales technology optimization, dashboards, and analytics.

↑ Back to top

#