Couplr AI research
The Find an Advisor Index
Can someone actually find the right advisor on your website?
We walked 358 of these pages by hand across the United States, Canada and South Africa and tested every one against six things a consumer should be able to do. Start with your own.
Free, and nothing is collected: no sign-up, no email, no cookie wall. Everything this panel does happens in your browser.
The industry is not bad at connecting you to an advisor. It is bad at letting you pick one.
We walked 358 “find an advisor” experiences the way a consumer would: no logins, no industry contacts, no phone calls. We tested each against six things a consumer should be able to do, and every one of the six is a fact you can check yourself in about ten minutes.
The result was not the one we expected. Financial services is no worse than therapists, doctors and lawyers at letting you reach a named person once you have found one. On two of the six tests it is slightly better. Where it collapses is the part that decides whether you meet the right person: 29% of surfaces let a consumer narrow by what they need, and 21% by who fits them. Outside financial services those numbers are 89% and 78%.
110 of 340 surfaces hand a consumer a named advisor they never got to choose. 24 do the opposite. The failure runs almost five to one in one direction.
34 surfaces out of 340 passed all six. 82 passed none.
This is not a new kind of study. In 2023 the US Senate Committee on Finance ran a secret-shopper audit of mental-health provider directories — 120 calls across 12 health plans in six states. A third of the listings were wrong, unreachable, or never called back. More than 80% of the listed in-network providers were, in the committee’s word, ghosts. Staff could book an appointment 18% of the time. We did the same thing to financial advice.
US Senate Committee on Finance, Behavioral Health Care When Americans Need It Most — secret shopper study, May 2023.
The standard
Six things a consumer should be able to do
There is no score on this page, and no ranking. Each surface is tested against six things, and each one is a fact you can check yourself in about ten minutes. The average financial surface passes 3.05 of six. The average surface outside financial services passes 4.61.
| The test | Financial340 | Benchmark18 |
|---|---|---|
| See a real person at allA named individual advisor appears at some depth without submitting a form. | 76% | 100% |
| Narrow by what you needA control for specialty, credentials, or how they are paid. Location alone never counts. | 29% | 89% |
| Narrow by who fits youA control for something about the consumer: the language they speak, the community they belong to, or a life event or life stage they are facing. | 21% | 78% |
| Reach them yourselfYou can get that named person’s own phone or email, or book into their calendar, without submitting your details to the firm first. | 60%financial ahead | 50% |
| Keep your details until you chooseThe route to them is not a queue, a callback, a “we’ll match you” form, a message to the firm, or someone else’s calendar. | 60%financial ahead | 56% |
| Be found by a machineThe advisor is visible and addressable in the served bytes, cookieless. | 60% | 89% |
Read the middle two rows and then the two below them. Financial services is three times worse than the benchmark at letting a consumer narrow, and no worse at all at letting them make contact. The industry has built the second half of the journey and skipped the first.
The benchmark is 18 surfaces. On tests 4 and 5 the two populations are within a few points of each other, and two benchmark rows moving would close the gap — which is why this page says financial services matches the benchmark there rather than beats it. The gap on tests 2 and 3 is far too wide to move that way.
How many of the six each surface passes
- 0 of 68224%
- 1 of 682%
- 2 of 6257%
- 3 of 63510%
- 4 of 611032%
- 5 of 64614%
- 6 of 63410%
340 financial surfaces. 34 pass all six. 82 pass none.
The commonest result by a distance is four of six, and it is the same four almost every time. Of the 110 surfaces at four, 88% are missing “narrow by who fits you” and 85% “narrow by what you need”. A surface at four of six has almost always built everything except the choosing.
The 34 that pass all six
AAAFA / Quad-A · American Family · BMO Private Wealth locator (CA) · Catholic Financial Life · CFP Board · CIBC · Coastal Community PWG · Coldstream · Edward Jones · Farmers · Farther · Focus Partners (Buckingham) · FP Canada · Garrett Planning Network · iA Financial · Manulife Wealth · Mariner Wealth Advisors · MassMutual (incl. MML) · Merrill · Modera Wealth · Morgan Stanley · Mutual of Omaha · New York Life · RBC Insurance · Savvy Wealth · SchoolsFirst FCU · State Farm · TD (advisor-match) · Thrivent · Waverly Advisors · Wealthtender · WiserAdvisor · XY Planning Network · Your Advisor Guide
Wirehouses, carriers, credit unions, professional bodies, marketplaces and startups. Whatever your size and whatever your regulator, somebody like you already does this.
What is this costing you?
The Six measure what a consumer can do on your page. They don’t measure how many actually get through.
Almost nobody measures that. Your marketing budget — paid search, organic, email, events, all of it — buys traffic to your website. Your find-an-advisor page is the last step before a human. Here is the arithmetic.
In one engagement we know the numbers for, a large wealth management firm’s find-an-advisor tool — linked from a homepage receiving over a million visits a year — produced about a hundred enquiries in a year. We are not publishing that as a rate, because we cannot confirm whether the traffic figure counted visits or raw server hits, and the two differ by more than an order of magnitude. What survives the ambiguity is the shape of it, and the question. You can answer it about yourself in about thirty seconds.
For context, on what that traffic is worth. Across the industry, a client acquired through marketing costs $3,119 on average — and 83% of that is not hard dollars at all, it is the imputed cost of the advisor’s own time ($2,600, against $519 of actual spend). Among advisors who use each strategy, the median runs from $390 for a client referral to $12,088 through a marketing consultant. The single cheapest activity measured is search engine optimisation, at $240 per client — $338 once you include the advisors it produced nothing for.
Kitces Research, How Financial Planners Actually Market Their Services, 2019. Figures among advisors using each strategy unless stated. Your website is the destination of most of that spend, whichever channel it went through.
The calculator is interactive and is omitted from the printed edition. Use it at couplr.ai/find-an-advisor-index.
What your steps bought
Two steps is not a good experience. It is a fast one.
A step is only worth taking if it buys the consumer something. So we never scored the number of steps on its own — we recorded what each surface’s steps actually bought. 86 surfaces are short because the consumer never got to choose, against 70 that are short and let them choose. The same two-step journey can be the best result on this page or one of the worst.
| Route | Financial340 | Benchmark18 |
|---|---|---|
| Short, and you choseTwo steps or fewer, and a control let the consumer narrow before they arrived. | 7021% | 633% |
| Longer, but the steps earned itThree steps or more, but the consumer got to choose on the way — the steps bought something. | 247% | 422% |
| Short — because you never got to chooseTwo steps or fewer, because nothing asked the consumer what they needed. Short is not the same as good. | 8625% | 00% |
| Longer, and the steps bought you nothingThree steps or more with no choice in them. The consumer paid in effort and got no say. | 247% | 00% |
| You can see them. You cannot reach them.A named advisor is visible, and there is no way to reach that person without going through the firm first. | 5416% | 844% |
| No named person at any depthNo named individual at any depth. | 8224% | 00% |
94 surfaces let a consumer choose on the way through. 110 moved them along without ever asking what they needed. The rest never produced a person they could reach at all.
Those 110 are the same 110 surfaces as the headline figure, arrived at from the other side. A route that completes is a route that reached somebody; a route with no choice in it reached somebody the consumer never picked. The two sentences describe one set.
Compare me to my own kind
Country is the wrong peer group for most readers
A credit union does not care how it compares with a wirehouse. Every surface carries a segment, applied identically across all three countries, so a firm can be set beside its own business model. This is a count of facts by segment. It is not a league table, and there are no positions in it.
| Segment | n | Mean of six | Median | By need | By fit | Machines |
|---|---|---|---|---|---|---|
| Cross-industry benchmark | 18 | 4.61 | — | 89% | 78% | 89% |
| Professional body registers | 15 | 4.60 | 5 | 87% | 67% | 80% |
| Matching operators n<5 | 1 | 4.00 | 4 | 100% | 100% | 100% |
| RIAs & independent wealth managers | 83 | 3.72 | 4 | 40% | 23% | 87% |
| Broker-dealers, IBDs & platforms | 58 | 3.31 | 4 | 9% | 16% | 64% |
| Banks & credit unions | 67 | 3.28 | 4 | 33% | 18% | 61% |
| Fraternal benefit societies | 17 | 2.41 | 3 | 12% | 12% | 41% |
| Lead marketplaces | 18 | 2.22 | 2 | 28% | 17% | 56% |
| Insurance carriers | 71 | 2.21 | 1 | 23% | 20% | 31% |
| Digital advice & D2C | 10 | 0.50 | 0 | 0% | 0% | 20% |
The two ends of that table are the finding. The professional bodies — non-profit, no lead-generation incentive, no reason to hold a consumer’s details — average 4.60 of six, above every commercial segment and level with the cross-industry benchmark at 4.61. Digital advice, the segment that exists because the old way was too hard, is the floor at 0.50. The barrier is not technical: the registers at the top of this table are, for the most part, plain pages with a search box. And it is not regulation on its own — 28 of the 34 surfaces that pass all six are commercial businesses operating under exactly those rules, among them 8 insurance carriers and 5 banks. What the top of this table has in common is the absence of a lead-generation incentive. That is our reading of the pattern, not something these six tests measured.
One thing passing all six does not require: giving up control of who is eligible. A firm can decide which advisors appear — by channel, by licensing, by capacity — and still let a consumer see, choose and reach the named people inside that set. The six measure named people and direct access. They say nothing about how large the roster has to be.
The businesses whose entire product is matching are the ones that do not match.
Of the 18 lead marketplaces, 13 let a consumer narrow by neither what they need nor who fits them — and 8 pass none of the six at all. 12 take the consumer’s details before they have chosen anybody. They are not matching services. They are distribution, and the matching is something that happens to the consumer rather than something they do.
The most technologically sophisticated segment is the worst at reaching a human.
The 10 digital advice and direct-to-consumer surfaces pass 0.50 of the six between them, and 7 of 10 produce no named person at any depth. That is a design choice, not a defect — but a consumer who decides they want to talk to somebody has nowhere to go.
Insurance carriers have the most people and the fewest ways to reach them.
34 of 71 carrier surfaces offer no route to a named advisor — the highest count of any segment operating a real advisor force.
Every expanded row in the Index carries its own peer line — a count of how many surfaces of the same kind clear the test that row failed. It is a count, never a position, and it is suppressed entirely for any segment with fewer than 5 surfaces in the comparison set, and for the cross-industry benchmark, which is not a peer group.
The comparison set — n = 18
Outside financial services, the failure runs the other way.
We ran the same walk on 18 non-financial surfaces where consumers choose a professional — therapy, medicine, law, tax and trades. They pass 4.61 of the six on average against financial services’ 3.05, which is the easy comparison and the least interesting one.
This is the interesting one. 17 of 18 benchmark surfaces let a consumer narrow before they choose, and only 9 of 18 let them reach the person afterwards. Financial services is the mirror image: 117 of 340 let you narrow, 203 of 340 let you make contact. Doctors and lawyers let you pick and then stand between you and the person. Finance hands you a person you never picked.
These 18 are never counted in a headline figure or a country section. Switch “Benchmark” on in the Index to read them alongside everything else.
The sixth test
Can a machine find your advisors?
Consumers increasingly ask an assistant before they ask a search engine. The sixth test asks whether an advisor is visible and addressable in the bytes the server actually sends — no cookies, no script execution, no login. 204 of 340 — 60% manage it.
| Machines can read | Surfaces | Share |
|---|---|---|
| Yes | 204 | 60% |
| Partly | 11 | 3% |
| No | 125 | 37% |
Independent RIAs are the most machine-visible segment in financial services by a wide margin at 87%. A roster rendered by script after the page loads is invisible to most of what now reads the web on a consumer’s behalf, and that is a publishing decision rather than a budget.
Side by side
Compare any two firms on the six
Compare two surfaces
Side by side on the six. This is not a ranking — it shows what a consumer could do on each, and nothing about which firm is better.
The ones with nowhere else to go
Matching operators
One surface in the study, which is too few to be a group. There is no page for a category of this size and no mean worth quoting on it — a mean on one surface is a single firm wearing a category’s name. The row is listed in full, here, so nothing in this study goes unpublished.
| Surface | The Six | Steps | Choose by | Ended in | Machines |
|---|---|---|---|---|---|
| You can see them. You cannot reach them.1 | |||||
✓✓✓✗✗✓
| —Seen, not reachable | Identity/community · Life event · Life stage +3 more | A calendar — but not that advisor’s | Yes | |
This walk ended in a calendar — but not that advisor’s. The six, and what carries each
What would change this row
| |||||
Every part of this report
By business model
Using this
Free to read, free to quote, free to republish
This report is published under a Creative Commons Attribution 4.0 licence. Quote it, chart it, republish it, put it in a deck or a filing. No permission needed and nothing to sign — we ask only that you name it and link to it, so a reader can check the figure against the row it came from.
Cite it like this
Couplr AI. The Find an Advisor Index: an audit of 358 “find an advisor” experiences across the United States, Canada and South Africa. Fieldwork closed 11 September 2026. https://www.couplr.ai/find-an-advisor-index/
Every row is on the page it belongs to, with its six results, the page we walked and the date we walked it. If you are checking one of our figures, those rows are the thing to check it against.
Method
Method, definitions and evidence
The Six, as tested
- See a real person at all
- A named individual advisor appears at some depth without submitting a form.
- Narrow by what you need
- A control for specialty, credentials, or how they are paid. Location alone never counts.
- Narrow by who fits you
- A control for something about the consumer: the language they speak, the community they belong to, or a life event or life stage they are facing.
- Reach them yourself
- You can get that named person’s own phone or email, or book into their calendar, without submitting your details to the firm first.
- Keep your details until you choose
- The route to them is not a queue, a callback, a “we’ll match you” form, a message to the firm, or someone else’s calendar.
- Be found by a machine
- The advisor is visible and addressable in the served bytes, cookieless.
The six routes
A route is a two-by-two on the same threshold — short or long, chose or did not — plus the two dead ends. Steps are never scored on their own.
- Short, and you chose
- Two steps or fewer, and a control let the consumer narrow before they arrived.
- Longer, but the steps earned it
- Three steps or more, but the consumer got to choose on the way — the steps bought something.
- Short — because you never got to choose
- Two steps or fewer, because nothing asked the consumer what they needed. Short is not the same as good.
- Longer, and the steps bought you nothing
- Three steps or more with no choice in them. The consumer paid in effort and got no say.
- You can see them. You cannot reach them.
- A named advisor is visible, and there is no way to reach that person without going through the firm first.
- No named person at any depth
- No named individual at any depth.
How we walked
One consumer, no account, no login, no phone call, from each firm’s own find-an-advisor entry point. We never submitted a form, booked a meeting, created an account, accepted a terms agreement, or entered personal data on any audited surface. A step is one discrete consumer action: loading a page is not a step; making a choice or a request is.
The numbers in this study, and what each one counts
Every count below is stated with what it is a count of, because a study whose strongest asset is publishing its own error rate cannot afford three unexplained denominators. 358 surfaces are walked, recorded and published here. 353 had been walked when the re-walk closed; 8 were added during the programme and have no earlier result to compare. 351 carry a comparable earlier result, and 199 of those changed — 57%. Of the 358 published, 340 are financial services, and that is the denominator for every percentage on this page; 18 are the cross-industry benchmark and are never counted in a headline. One further surface — Couplr’s own — was walked and is not published; the Disclosure says why.
| Count | n |
|---|---|
| Surfaces walked, recorded and published | 358 |
| Walked, but withheld because we publish it — see Disclosure | 1 |
| Of those, financial services — the denominator for every percentage on this page | 340 |
| Cross-industry benchmark, counted separately and never in a headline | 18 |
| Already walked when the re-walk closed | 353 |
| Added during the programme, with no earlier result to compare | 8 |
| Carrying a comparable earlier result | 351 |
| Of those, results that changed on the re-walk | 199 57% |
We walked it twice, and we got things wrong the first time
Every row was re-walked under a frozen protocol and forty-three written rulings. 199 of the 351 surfaces with a comparable earlier result changed — 57%. The errors were overwhelmingly ours, not the industry’s: dead and mis-registered URLs, rosters counted one page deep, probes pointed at the wrong page, controls invisible to a DOM scan.
The worst of them was an instrumentation defect. Browser tabs that were never painted did not render lazy-loaded advisor widgets, which manufactured error-free, HTTP-200, genuinely blank pages — surfaces that looked like total failures and were not. A sweep of all 67 rows that could carry that defect found 19 rendering false negatives, of which 4 were genuinely recorded wrongly. Two large banks moved from zero of six to six of six.
We publish this because no comparable study does, and because it is the reason these numbers can be defended. A study that cannot say what it got wrong has not finished checking.
What the Six do not measure
Advice quality, price, suitability, outcomes, or whether any advisor is any good. And they measure a floor, not a ceiling — a surface can pass all six and still be mediocre. One that passes all six serves a single language. We say so because a standard that states its own limits is far harder to attack than one claiming to measure everything.
They also do not measure what choice is worth. This index asks whether a consumer can find, choose and reach an advisor. It does not ask whether the relationships consumers chose outlast the ones firms assigned, or earn more of a household’s business. We have not found public data that tests it, or that separates a choice from a referral. The firms that track retention by lead source are holding the answer. If yours would contribute anonymised data to a follow-up study, tell us.
What test 3 counts, and what it does not
Test 3 counts controls that describe the consumer: the language they speak, the community they belong to, or the life event they are facing. It does not score, and this index does not recommend, filtering advisors by the advisor’s own race, religion, sex or any other protected characteristic. Where the cross-industry benchmark records such controls — several therapy directories carry them — that is an observation of what those sites offer, under a different regulatory context, not a standard we are holding financial firms to.
Where the answer depends on the reading
Test 5 has a stricter reading available, in which any wall that takes an email before a consumer can finish choosing disqualifies the surface. Under it, one benchmark row falls by one and the gap between financial services and the benchmark narrows. Both readings are recorded on the rows. We publish the sensitivity rather than the single number that flatters the result, because a figure that moves under a defensible alternative reading should say so before somebody else says it for us.
What we did not claim
The strongest review of algorithmic matching in the adjacent literature finds that the access and communication functions of matching services are valuable while questionnaire-based compatibility prediction is not supported (Finkel, Eastwick, Karney, Reis & Sprecher, 2012). This study measures access. It makes no compatibility claim.
Sources
- US Senate Committee on Finance — Behavioral Health Care When Americans Need It Most (2023)
- RAND Corporation — Trust and Financial Advice (WR-1075, 2015)
- Kitces Research — How Financial Planners Actually Market Their Services (2019)
- Finkel, Eastwick, Karney, Reis & Sprecher — Online Dating: A Critical Analysis (2012)
Corrections and additions
Think we got you wrong? Not on the list at all?
Every finding here is an observation of a public page on a stated date. Pages change, and we make mistakes — we changed more than half of our own results before publishing. If your row is wrong, tell us and we will re-walk it. If your firm is not here at all and it has 25 or more advisors, we will walk it. There is no charge and nothing to sign up for, either way.
How a re-walk works: we walk the page again under the same frozen standard, and we publish the new result with its date, beside the old one. The original stays visible. A re-walk changes a score only because the page changed — never because of who is asking, and never faster for a Couplr customer than for anyone else.
Submit a correction at couplr.ai/find-an-advisor-index
Corrections log
Most recent first. Started on the day of publication with the corrections this study made to itself.
- 11 September 2026The whole index — Every row was re-walked under a frozen protocol before publication. More than half of all results changed. The errors were overwhelmingly ours — dead and mis-registered URLs, rosters counted one page deep, probes pointed at the wrong page.
- 11 September 2026Sixty-seven rows, re-swept — An instrumentation defect meant unpainted browser tabs never rendered lazy-loaded advisor widgets, manufacturing error-free, HTTP-200, genuinely blank pages. A sweep of all 67 rows that could carry the defect found 19 rendering false negatives, of which 4 were genuinely recorded wrongly. Two large banks moved from zero of six to six of six.
- 11 September 2026The Canadian language finding — An earlier figure counted a language filter that has never worked, and has been withdrawn. What replaced it is the stronger finding: several Canadian surfaces advertise a language control that the page never transmits.
- 11 September 2026The standard itself — The five-band rating used in earlier drafts was retired. It compressed six checkable facts into one number nobody could audit. The Six (v2.2) replaced it, and this page carries no score of any kind.
- 2 October 2026FPA PlannerSearch, re-walked — The Financial Planning Association rebuilt its planner search after our walk. The open roster that let a consumer see and contact a named planner has been replaced by a ten-step questionnaire, and planners’ own telephone numbers have been removed. On the re-walk the surface passes four of the six rather than all six: a consumer can still narrow, and can still send a message, but can no longer reach a named planner directly or keep their details until they have chosen. This is a change at the firm, not an error of ours — it is logged because a published finding moved. With it, the professional body registers stopped sitting above the cross-industry benchmark and became level with it.
Disclosure
Who published this, and what they sell
This study is published by Couplr AI, which sells software that matches consumers to advisors. That is a commercial interest in the problem this study describes, and you should read it knowing that. Three things follow from it.
No Couplr statistic appears anywhere in this report. The argument is carried by the audit and by independent research. Every place Couplr speaks for itself is labelled as Couplr speaking, and there are three of them.
Couplr is not in the index. We publish this study and we sell software in this category, so a finding we wrote about ourselves would not tell a reader anything, and any figure computed with our own row in it would be a figure computed on the publisher.
The rule is the rule, not a convenience: we also excluded match.tacdesignee.com, which Couplr powers for The American College. We do not publish findings about surfaces we built, whoever they belong to. Every figure in this report is computed on the 340 financial surfaces that are left.
One surface was excluded for safety. It is named rather than quietly dropped, because a reader checking our coverage should be able to see what is missing and why.
eternity.co.za— Redirects to a fake-CAPTCHA malware download page. Excluded from the corpus; never fetched, never followed, and never rendered as a live link. It is named here as excluded rather than silently dropped.
Seven candidates were investigated and not admitted. None of them runs a consumer advisor-finding surface at any depth, so each was struck rather than counted as a failure — a firm that never made the claim is not a firm that failed it, and counting it as one would flatter every percentage on this page. They are named here, with what the enumeration found, so the decision can be checked. Two of them were in this index until 2026-09-22 and were reconsidered when the admission test was tightened. They are named here rather than quietly dropped.
- Invesco
An asset manager that reaches consumers through third-party advisors rather than employing any. Its entire consumer namespace is three pages – landing, account overview, contact – and it operates no advisor-finding surface at any depth: 199 nav links enumerated with mega-menus expanded, 1,900 sitemap URLs across three sitemaps, five canonical locator paths all 404, 34 locator vendors absent. - Envestnet
A business-to-business wealth-technology platform whose customers are advisory firms rather than consumers. Its own served bytes state, on every page: “This website is for investment professionals only. It is not intended for private investors.” That is a declared refusal, not our inference. No consumer contact route exists at all – zero phone numbers, zero emails; Contact Us resolves to a demo request. - goldmansachs.com
The parent front door standing above the host where the registered private-wealth product lives – Goldman appends its own referral parameters to every link it sends into that host. Complete sitemap enumeration, 6,911 URLs: nine advisor-shaped URLs, all read and none a locator. No second consumer advisor-finding surface. Registering it would put two ids over one roster. - Global Atlantic
Counted in this index until 22 September, then reconsidered and struck when the admission test was tightened. It sells annuities through independent distribution and supplies no advisor of its own: its instruction to the consumer, on every product page, is “Talk to your financial professional about ForeIncome today” – an advisor the consumer is assumed to already have. Proven three independent ways from the firm’s own material: all 70 consumer navigation links enumerated, the only professional route gated “This site is intended for financial professionals only”; a complete 98-URL sitemap with no advisor, agent, producer, team or profile page; and a CMS whose own content-type facets across 193 items contain no person-shaped type at all. - STANLIB
Counted in this index until 22 September, then reconsidered and struck when the admission test was tightened. An asset manager whose customer is largely the adviser. The site has four front doors and the second reads “I am a financial adviser”, leading to fund materials, forms and portfolio tools for advisers to use with their clients. There is no door that says “I need one”. Every consumer route ends at one share-call number or a Johannesburg street address; eleven people are named on the whole site, all of them the executive leadership team; and the firm’s own 672 KB site bundle contains no adviserName, no locator, no province, no postalCode and no language. The adviser is not an entity this website models. - USA Financial
A business-to-business broker-dealer, TAMP and practice-management platform whose customer is the independent advisor, not the consumer. It operates no advisor-finding surface at any depth: complete sitemap enumeration, 462 URLs, not one a locator, and robots.txt declares no second sitemap. The only roster on the domain is 55 corporate staff – Chief Financial Officer, Human Resource Generalist, Accounting Associate, IT Director – and forecloses it. Its own contact form offers the consumer one option, “an Investor with assets at USA Financial”, which is an account-service path for an existing client rather than a way to choose an advisor. DJ, 22 Sep: struck rather than counted as a failure, because a firm with no find-an-advisor experience at all is outside what this index measures. DISCLOSED: USA Financial is a Couplr commercial prospect; the exclusion is on the same ground as Invesco and Envestnet and predates any commercial discussion of the index. - vanguard.com / investor.vanguard.com/corporate-portal
One page, not two: vanguard.com lands on the corporate portal, which is an audience switcher with six doors, zero advisors and zero input elements – labelled “More Vanguard sites” in the firm own retail footer. Its four advice tiers ($100 / $50,000 / $500,000 / $5M+) are distinct products under, but NONE names an individual, so none is a second advisor-finding surface. Complete sitemap enumeration, 859 URLs: no directory, no profile page, no locator.
We published our own errors at length — in the method section and in the corrections log, including the instrumentation defect that made working pages look broken. Every row behind these figures is on this page: the six results, the page we walked and the date we walked it. Any number here can be checked against the rows it came from.
Published under CC BY 4.0 — free to quote and republish with attribution to Couplr AI. Firms are named in this index; individual advisors are not, anywhere.