Research
Seven things we keep seeing in advisor transitions and client trust, for the firm leaders and advisors who are living them, and what we found when we went looking for the evidence behind them. Including where it pushed back.
Then everything we've written, the reading behind it, and the show.
Every one of these came out of our own writing before we went looking for evidence. Where the research supports it, we say so. Where it pushes back, we say that too.
Documentation transfers. Trust doesn't.
A founder retires, a successor steps in, a firm completes an acquisition. Accounts move, forms update, the meeting gets scheduled. Inside the firm the transition looks complete. From the client's side the relationship may still be undecided, because what moved was the file and not the understanding.
Advisor Change Can Look Complete on Paper · The Plan Transfers, The Understanding Doesn't
What the evidence says
Cerulli finds advisors who move firms lose 11% to 22% of the assets they managed. McKinsey finds 32% of affluent investors switch firms when their advisor leaves, and expects roughly 110,000 advisors to retire this decade. Capgemini finds 81% of inheritors plan to switch within one to two years. The handoff is where the industry loses people, and it is scheduled to happen more often.
The share that never arrives: 11% to 22% of assets, by Cerulli's count
Trust drifts before it breaks, and the money moves before anyone says anything.
Trust rarely fails in one dramatic moment. It starts with a slow reply, a spouse's concern that gets missed, a detail someone expected to be remembered. The client keeps their accounts in place and stops bringing new assets. No complaint, no transfer request, nothing to log. Firms measure the visible transition and miss the felt one.
The Moment Trust Starts Drifting · McKinsey Identified the Advisor Shortage
What the evidence says
We wrote about this before we went looking for data on it. Schwab's RIA Benchmarking Study has reported client retention holding steady at 97% for a decade, and an average that stable can hide what happens at the specific moments that test a relationship. The research supports the shape of it. PriceMetrix holds the industry's only large behavioural retention dataset, roughly 7 million investors, and every variable that predicts retention in it is structural: account types, tenure, fee basis. Conversation quality isn't in the model because the data doesn't contain it. Meanwhile the top decile of advisors retain 98% of clients and the bottom decile 84%, a fourteen-point spread the structural variables don't explain. Retention numbers move late and move little. Net new assets shows the change first.
Retention holds while net new assets moves first
Clients don't tell you what is wrong. Often they can't.
They rarely say "I feel less important now" or "I'm not sure this new advisor understands us." They ask practical questions instead. Who do I call, will anything change, when is the next meeting. The practical question is often carrying a deeper one, and the first signal of risk isn't a complaint. It is hesitation, a quieter spouse, a shorter meeting, a client who says everything is fine.
What Clients Feel but Don't Say During Advisor Transitions
What the evidence says
The strongest support on this page, and it isn't from our industry. Bond, Carlson and Keeney asked people to list what mattered to them in a real decision. They named 5.9 objectives on average and recognised 14.3 when shown a list, rating the missed ones as nearly as important as the ones they thought of. Given a full week to deliberate, they still missed more than half. People aren't withholding. They can't retrieve it on demand, and no amount of asking nicely fixes that.
Bond, Carlson and Keeney: what people name unprompted, against what they recognise from a list
Four more we keep seeing.
"Personal" means something different to the client than it does to the advisor.
Advisors tend to hear personal as warmth.
The evidence. Two independent findings arrived at this after we published it, and both are sharper than we were.
Feel More Personal
A perfect transcript still would not tell you who you are sitting with.
Hand a new advisor every word a client ever said and they still don't know the person, because the meaning was never in the words.
The evidence. The Oasis Group compared six advisor notetakers and found note accuracy between 85.9% and 96.15%.
Capturing What a Client Says Isn't the Same as Knowing Them · Advisors Remember Their Clients, But Nothing Connects What They Know
Steadiness comes before solutions, and it isn't the soft part.
Advisors rarely lose trust by giving the wrong answer.
The evidence. Zolnierek and DiMatteo pooled 21 experimental studies covering 1,280 physicians and 10,190 patients and found that training the professional in how to conduct the conversation raised the odds of the patient following through by 1.62 times.
Steadiness Before Solutions
The information advantage is going. What's left is judgment.
Advisors used to hold what clients could not get: research tools, technical knowledge, access, experience with situations the client had never faced.
The evidence. Russell Investments puts advisor value at 4.92% with behavioural coaching the single largest component at 2.30%, and frames the shift directly: "The differentiator is no longer access to information, but the ability to apply it with judgment, context and discipline." Brynjolfsson, Li and Raymond, in a field experiment with 5,172 support agents, found AI assistance raised productivity 15% overall and about 30% for the least experienced, compressing the experience curve rather than replacing the expert.
Your Next-Gen Clients Are Already Coming With an AI Plan · The Judgment Age · AI Is Not Just Coming for Tasks
Every piece is on this site in full. All the writing, in one place →
Series · The Rise of Human Wealth Tech
Four parts on how the industry is finally building around the person, not just the portfolio.
The Financial Side Runs on Software. The Human Side Runs on the Advisor.
Names the shift. Every technical part of advice now has software behind it. The human part still runs on one person's memory and attention, which is the only part of the stack that has never been built.
The Plan Transfers. The Understanding Doesn't.
What it costs when a household changes hands. The understanding that makes the relationship work lives in the advisor rather than the firm, so when the advisor leaves it walks out with them, which is why firms lose a share of clients at every transition.
Capturing What a Client Says Isn't the Same as Knowing Them
Takes on the fix most firms reach for first. Better notes, fuller transcripts, record everything. A transcript is a record of words and understanding is what those words mean. Nobody writes down a pattern they have not noticed yet, and most of what an experienced advisor knows, they know without knowing they know it.
What Advisors Carry in Their Heads, a Firm Can Now Read
Our answer, and the first public description of what we built. Sets the test a real solution has to pass before naming anything we make.
Series · Trust at Transitions
Four parts on how wealth firms keep clients through the moments that test trust.
What Brings Clients Back Is Real, But What Keeps Them Isn't Built Yet
Starts from Ari Galper's observation that clients who leave and return are chasing the memory of being understood. Agrees with him, then asks the question he leaves open: if early attention is what builds trust, why does it fade so reliably even with advisors who genuinely care? Because it is a structural problem, not a character one.
Onboarding Isn't the Problem, But What Happens After It Is
Onboarding is the high water mark of almost every advisor relationship, the one moment the whole process is oriented toward the person rather than the plan. Then the annual review rhythm takes over and the agenda defaults to allocation and performance. Names the failure path: discovery to drift.
Advisors Remember Their Clients, But Nothing Connects What They Know
Goes inside the workflow. CRMs track transactions and notes track what was said. Nothing tracks what it meant. The daughter's college plans, the spouse who seemed checked out, the comment about retiring two years early. Context decays, and neither side can name when the relationship started feeling managed instead of known.
Trust Was Always the Point, But Nothing Was Built to Hold It
The vision piece. What the advisor-client relationship requires at the infrastructure level rather than the behavioural one, and what it is worth to a firm that builds it.
On advisor change
Advisor Change Can Look Complete on Paper While Trust Is Still Quietly Undecided
Accounts transferred, forms updated, meeting scheduled. The client is still asking whether the new advisor knows them, knows their spouse, and knows what matters now. Documentation transferred. Trust may not have.
The Moment Trust Starts Drifting
Trust rarely breaks in one dramatic moment. It starts with a slow reply, a missed concern, a detail nobody remembered. The financial signal often arrives before anyone names the relational one: the client keeps their accounts and stops bringing new assets.
What Clients Feel but Don't Say During Advisor Transitions
Clients rarely announce the emotional layer. They ask who to call and when the next meeting is, while carrying questions they never voice. The first signal of risk is hesitation rather than complaint.
Advisors hear "personal" as warmth and remembered birthdays. Clients often mean something else: I don't have to re-teach you who we are. Two research findings have since confirmed the distinction, and one of them found the advisor's version can actively cost trust.
On the human side of advice
Most advisors don't lose trust by giving the wrong answer. They lose it because the client wasn't steady enough to receive it. Orientation first isn't the soft part of the work, it is what makes the technical part land.
Most adults face 30 to 40 disruptors in a life, and three to five become lifequakes: the moments that shake identity, confidence, relationships and decision-making. Major disruption isn't rare. It is a feature of the client lifecycle.
The Judgment Age: Why Next-Gen Clients Still Need Human Advisors in an AI-Shaped World
Starts from Karpathy's job exposure visualiser and the observation that most of the advisor workflow is becoming digital. Next-gen clients are digital-first but not human-last. They want advice that feels different, not less advice.
AI Is Not Just Coming for Tasks. It Is Coming for Your Identity.
For years professionals built their value on what they knew. The harder question underneath the job disruption conversation is what you become when a machine can produce the thing that made you feel valuable.
Your Next-Gen Clients Are Already Coming to Meetings With an AI Plan
Built on Brian Portnoy's distinction between the advisor as mechanic and the advisor as guide. Clients now arrive with the mechanic work already done. If information is no longer scarce, what is the advisor for?
On the industry
Reads the McKinsey shortage numbers as a retention problem rather than a staffing one, and lands on the 32% who switch when their advisor leaves. Sets the average against the moment: Schwab reports RIA client retention holding at 97% for a decade, and an average that stable can hide what happens at succession, reassignment and acquisition.
The Missing Layer in the Wealth Management AI Conversation
On the question that never gets asked in an AI demo: does the system actually know what is going on with this client?
Why the Fastest-Growing Wealth Firms May Be Creating Their Biggest Retention Problem
Growth creates handoffs, and handoffs are where relationships get tested. The firms compounding fastest are manufacturing the exact moment their systems are least built for.
Grouped by what it tells you. Every entry links to the publisher. Where a report is restricted to financial professionals, or paid for by someone with an interest in the answer, we say so. Where a finding complicates what we think, it sits in its group alongside everything else.
The advisor shortage, and what it does to clients
The looming advisor shortage in US wealth management
The industry could be short 90,000 to 110,000 advisors by 2034. Roughly 110,000 advisors, 38% of the workforce holding about 42% of industry assets, are expected to retire this decade.
The finding most people miss sits one line further on: 32% of affluent and high-net-worth investors switch firms when their advisor leaves. A staffing problem and a retention problem are the same problem.
Advisor retirement and transition research
Advisors who move firms typically lose 11% to 22% of the assets they managed, depending on channel. Roughly 40% of advisory assets are expected to transition as advisors retire over the next decade.
Cerulli arrives at nearly the same place as McKinsey by a different route, which is the strongest kind of corroboration.
Canada, where the numbers thin out
Fewer Canadians are working with an advisor
61% of investors work with a financial advisor, down eight points from 2020, with the steepest decline among investors under 45 and those with less than $100,000 invested. 45% now hold a self-directed account, and 30% of those opened it in the previous two years.
A national sample of 7,215 adults weighted to Census data, and the longest-running Canadian investor series we could find.
Canadian advisors on their own succession
31% of Canadian advisors plan to retire within ten years and 44% have no succession plan of any kind. 52% rate their firm's support for managing client transitions as fair or worse.
Sponsored by IG Wealth. Methodology disclosed, sample weighted.
Where Canadian wealth is going
Canadian household investable assets are projected to reach C$11.2 trillion by 2032, with households over 55 holding more than C$7 trillion of that, or 63% of the total. Generation X is forecast to receive roughly C$400 billion in inheritance flows over the decade.
The most serious Canadian transfer forecast we could find. If you have seen a "$1 trillion between generations" figure quoted, see the last group on this page.
What clients say, and what they do
What clients praise when they praise their advisor
Across more than 2,500 client reviews, 89% of what clients praise has nothing to do with investment performance. It is about feeling heard, supported, and understood.
The inheritance problem, and what advisors say they need
81% of inheritors plan to switch firms within one to two years of inheriting. Asked what capability matters most for serving them, advisors ranked a holistic view of the client and insight they can act on above automating meeting summaries and emails.
The second half is the part worth sitting with. Advisors are asking for understanding, not more automation.
The conversations that aren't happening
51% of advised investors under 40, and 39% over 40, say their advisor has discussed what a future wealth transfer will require. 18% say their advisor has met with or offered to meet other family members.
More useful than any prediction about heirs firing advisors, because it measures what is being done rather than what someone expects to happen.
Why do investors fire their financial advisor?
Of the reasons investors gave in their own words, quality of advice and services accounted for 32% and quality of the relationship 21%, against cost at 17% and investment performance at 11%.
The base matters and we will not quote those percentages without it: 3,003 people surveyed, 185 who had ever fired an advisor, 184 codeable answers. Morningstar's own headline is that firing is rare, which is a finding in its own right and one that complicates the industry's retention anxiety.
Most widows stay, and the industry keeps saying otherwise
Cerulli found 85% of widows and widowers remained with the incumbent advisor after a spouse's death. Kehrer Group and RFI Global, analysing a separate database, put widow departure at about 14% against a 5% baseline across all investing households.
Two datasets, the same answer. Widows are roughly three times more likely to leave than the average client, and the overwhelming majority stay. The transition risk is real and it isn't the catastrophe the industry describes.
What is actually known about the conversation
Can decision makers articulate what they want?
Asked to list what mattered to them in a real decision, people named an average of 5.9 objectives. Shown a master list afterward, they recognised 14.3 as personally relevant, and rated the ones they had missed as nearly as important as the ones they thought of. Given a full week to deliberate, they still missed more than half.
If you read one thing on this page, read this. It is the reason good discovery is hard, and it has nothing to do with whether the advisor is asking in good faith. Almost nobody in this industry cites it.
Up to 75% of people changed at least one of their top three financial goals after going through a structured process, moving from surface goals such as retirement or a house toward underlying motivations.
The financial replication of the finding above. Sample sizes aren't stated in the paper, which is a real limitation. Vendor research.
Does preparing the professional actually change anything?
Across 21 experimental studies covering 1,280 physicians and 10,190 patients, training the professional in how to conduct the conversation raised the odds of the patient following through by 1.62 times.
The only causal evidence we could find, in any field, that equipping the professional changes what the client subsequently does. We quote the experimental figure rather than the larger correlational one that is also available. This is medicine rather than advice, and the effect is a nudge rather than a transformation.
Specificity reads as listening, until it doesn't
In real service conversations, more concrete and specific language predicted higher satisfaction and higher customer spending over the following 90 days, with perceived listening as the mechanism. When the detail wasn't relevant to what the customer had actually raised, concreteness reduced perceived listening.
The boundary condition is the part that matters. Knowing more about someone only helps if you surface the right part of it, which is a caution that applies to us as much as to anyone building in this space.
What communication is actually associated with
Among 1,088 planning clients, scheduled meetings were associated with higher satisfaction, trust and commitment, levelling off around four a year. Educational communication was associated with likelihood of continuing and with share of household assets. Communication about interests and hobbies, past a point, was associated with lower trust.
The closest thing financial advice has to evidence on this question, and it is a cross-sectional self-report survey from 2014. Assets are what the client says in broad bands rather than what the custodian records. We include it because it is the field's best attempt and because its last finding is one we had reached independently.
AI in advice, measured rather than promised
How fast AI notetakers actually arrived
The category went from one tracked product to fourteen in a single year, reaching 42.86% adoption among advisory practices. One of the fastest adoption curves ever recorded in advisor technology.
Artificial Intelligence Note Takers Research Report
An independent comparison of six advisor-specific notetakers. Note accuracy ranged from 85.9% to 96.15%, and action-item accuracy from 62.5% to 87.5%.
These tools work, and they work well. The most interesting result is a miss: none of the six caught an implied relationship action tied to an expected family event, because it was never said out loud. That gap is structural rather than a defect.
A field experiment with 5,172 support agents. AI assistance raised productivity 15% on average and about 30% for the least experienced workers, letting two-month hires match the performance of workers with six months or more.
Not a wealth management study, but the clearest evidence anywhere that this kind of assistance compresses the experience curve rather than replacing the expert.
What advice is worth
Estimates advisor value at 4.92%, with behavioural coaching the single largest component at 2.30%, nearly half the total. Not asset allocation, not tax, not product selection. Influencing behaviour.
Their framing of the shift is worth quoting directly: "The differentiator is no longer access to information, but the ability to apply it with judgment, context and discipline." Restricted to financial professionals. Link to Russell rather than hosting.
The emotional and time value of advice
86% of advised investors report more peace of mind, and three in four report saving time. At signup, 87% cited portfolio value as a reason and 74% cited emotional value. After enrolling, 86% reported experiencing emotional benefits.
The gap between why people sign up and what they report afterward cuts both ways. Clients buy portfolio management. The relationship is what they discover later. Vendor survey of its own clients.
The value of financial advisors, in Canada
Between 2010 and 2014, Canadian households that kept their advisor saw assets rise 16.4%. Households that dropped their advisor saw 1.7%.
The only Canadian econometric work we could find on this question, and what it isolates is continuity of the relationship rather than quality of the portfolio. Industry-sponsored research stream.
Numbers we stopped using
Follows every citation behind "70% of families lose their wealth by the second generation, 90% by the third" back to its source. All of them lead to one mid-1980s study of 200 family manufacturing businesses in Illinois, measuring whether majority family ownership passed to the next generation. The frequently cited MIT source restates the same study. The Economist article contains no such figure.
We link this because a page claiming a standard of evidence should be willing to apply it to the statistic the industry likes most.
Three more we checked and stopped repeating
"70% of widows leave their advisor," traced to an out-of-print report whose original responses likely included people dropping a life insurance agent after a claim. The real figure is about 14%. "$1 trillion moving between Canadian generations," which traces to a 2016 CIBC note forecasting $750 billion over a window that has now closed. "95% of AI pilots fail," which is MIT Project NANDA reporting no measurable profit impact within roughly six months, on a base of 52 interviews and 153 responses, largely because the pilots had no baseline to measure against.
Some of these have appeared in our own material. Correcting them in public costs less than being caught using them.
Advising Humans is mostly the two of us thinking through the human side of advice, with playbook episodes on the transitions that test a relationship hardest: divorce, business exit, retirement, career change, advisor change.
It is where a lot of this argument got tested before it was written down. The episodes aren't interviews with a rotating cast of guests, and we would rather say so than imply a research programme we don't have.
If any of this sounds like your firm, and you'd rather do something about it than read more, we're running a small preview with a handful of firm leaders. Request a preview invite →
