Which Households Will You Actually Lose? A Pre-Move Attrition Risk Score for a Book in Transition
Aggregate retention benchmarks describe a book, not a household. Before a move, score each household on six observable factors: relationship depth, account complexity, tenure, life stage, the strength of the old firm's counter-relationship, and operational friction. The distribution tells you where to spend the outreach hours you actually have.
Every advisor planning a transition has been told a retention number. Ninety percent, ninety-five, sometimes higher. Those figures are contested for good reason, and we have written separately about why the headline attrition benchmarks by move type rarely mean what people think they mean.
But there is a more practical problem with a single retention percentage: it is useless for planning. Knowing you might lose 10 percent of the book tells you nothing about which 10 percent, and the entire value of knowing would be to prevent it. An advisor with 180 households and roughly 40 hours of client outreach capacity in the critical window needs a ranking, not an average.
This is a rubric for building that ranking before you move.
Why the aggregate number fails operationally
Three reasons the headline figure cannot be worked with.
Attrition is concentrated, not distributed. Losses cluster in identifiable segments. A book does not lose an even slice off every household. It loses the accounts that were never really yours, plus a handful of relationships that were genuinely yours and were mishandled during the window.
The expensive losses are not the numerous ones. Ten small accounts leaving and one large multi-generational relationship leaving are the same headcount and wildly different outcomes. A household count treats them identically.
The window is short and capacity is fixed. Between resignation and full repaper there is a bounded number of meaningful client conversations one advisor can have. Spending them in account-size order is the default, and it is not the same as spending them in risk-adjusted order.
The goal of scoring is to answer one question: if I can have 40 real conversations in the next three weeks, which 40?
The six factors
Score each household on each factor, 0 to 3, where 3 means highest risk. The specific weights matter less than the discipline of scoring every household on the same axes.
1. Relationship depth (who owns the relationship)
The strongest predictor and the hardest to be honest about.
| Score | Signal |
|---|---|
| 0 | You brought the client in, they know you personally, no meaningful firm relationship |
| 1 | Long tenure with you, some firm touchpoints |
| 2 | Inherited from a departing or retired advisor at the firm |
| 3 | Firm-sourced lead, never met you outside a formal review, or primary contact is a firm employee who is not you |
Inherited households are the classic blind spot. An advisor who took over 30 accounts from a retiring colleague five years ago often counts them as theirs. The client may still think of the relationship as belonging to the firm, or to the predecessor. Those households do not follow at anything like the rate the advisor expects.
The honesty test: can you name the client's spouse, their children, and the reason they invest? If not for a given household, score it 2 or 3 regardless of how long it has been on the book.
2. Account complexity and operational friction
This is the factor most attrition models omit entirely, and it is the one that turns a willing client into a lost one.
| Score | Signal |
|---|---|
| 0 | One or two accounts, all ACATS-eligible, clean registration |
| 1 | Three or four accounts, standard types |
| 2 | Multiple accounts including a retirement plan, trust, or entity registration |
| 3 | Non-ACATS assets, annuities, alternatives, direct-held positions, or a known data problem |
The mechanism here is simple and under-discussed. A client who says yes and then spends six weeks signing forms, receiving confusing custodian mail, and watching positions sit in limbo is a client whose confidence in the decision decays daily. Every additional week of operational friction is another week in which the old firm's retention team has an opening.
That makes clean repapering a retention lever, not just an efficiency one. It is why the operational and relationship tracks cannot be planned separately, and why our client communication sequence mapped to the transfer timeline sequences outreach against the actual ACATS milestones rather than a generic calendar.
3. Tenure and stage in the relationship
| Score | Signal |
|---|---|
| 0 | More than seven years, multiple market cycles together |
| 1 | Three to seven years |
| 2 | One to three years |
| 3 | Under a year, or onboarded but never fully engaged |
Newer relationships are more fragile in both directions. A client who joined eight months ago has not yet accumulated the trust that survives disruption, and has recently demonstrated a willingness to change advisors, which is itself information.
4. Life stage and imminent events
| Score | Signal |
|---|---|
| 0 | Stable, accumulating, no known events |
| 1 | Approaching retirement, no immediate action |
| 2 | In distribution, or a liquidity event within 12 months |
| 3 | Estate settlement, divorce, business sale, or inheritance in progress |
A household in the middle of a life event is both the highest-risk and the highest-value to hold. These clients are actively re-evaluating everything, and a transition gives them a natural moment to reconsider the advisory relationship. They also cannot tolerate operational delay, because their event has its own deadlines.
Flag any household in this row for a personal conversation before the move rather than after, subject to what you are permitted to discuss at that stage.
5. The old firm's counter-relationship
| Score | Signal |
|---|---|
| 0 | No other firm touchpoints, no banking or lending relationship |
| 1 | Occasional firm service contact |
| 2 | Held-away products, a firm specialist relationship, or firm-branded loyalty |
| 3 | Mortgage, securities-based lending line, private banking, or a family member also serviced by the firm |
Lending is the sharpest version of this. A client with a securities-based line of credit or a mortgage tied to the firm has a practical reason to stay that has nothing to do with advice quality. These are not lost causes, but they require a specific plan for the credit relationship, and that plan needs to exist before the client asks.
It is also worth remembering that the old firm gets a formal opportunity to communicate. FINRA Rule 2273 requires the recruiting firm to deliver an educational communication to a transferring advisor's former customers, which prompts clients to ask specific questions about costs and assets that may not transfer. Households already inclined to hesitate will hesitate harder after reading it. Anticipate the questions rather than being surprised by them.
6. Engagement signal
| Score | Signal |
|---|---|
| 0 | Responsive, initiates contact, attends reviews |
| 1 | Responsive when contacted |
| 2 | Slow to respond, missed the last review |
| 3 | Has not engaged in 12 months, or contact information is stale |
The stale-contact case deserves its own note. A household you cannot reach in the transition window is functionally at maximum risk, because your entire retention approach depends on reaching them. Verify contact details on the whole book before the move, not during it.
Turning scores into a plan
Sum the six scores for a range of 0 to 18. Then cross the risk score against household value.
| Low risk (0 to 5) | Medium risk (6 to 11) | High risk (12 to 18) | |
|---|---|---|---|
| Top-value households | Confirm early, brief touch | Personal call in the first 48 hours, named owner | Pre-move planning, senior involvement, escalated ops handling |
| Mid-value households | Standard sequence | Personal call in week one | Personal call plus expedited paperwork |
| Long-tail households | Standard sequence | Standard sequence plus one call | Decide deliberately whether to pursue |
That bottom-right cell is the one nobody wants to talk about. Some households in a book are high risk, low value, and operationally complex, and pursuing all of them consumes capacity that would retain a top-value relationship instead. Deciding in advance which relationships you are not going to fight for is uncomfortable and it is better than making that decision accidentally by running out of hours.
The cell that most often gets under-served is top-value, medium-risk. These clients look fine on a size-ranked list, get a standard touch, and are the ones people are surprised to lose. Risk-adjusted ranking exists to surface exactly this group.
Building the score without guessing
Four of the six factors are judgment calls the advisor makes. Two are data, and the data half is where books go wrong.
Account complexity and engagement signal should come from records, not recollection. Specifically:
Pull the actual statements, not the CRM summary. The CRM tells you what someone entered. The statement tells you what the custodian holds: the real registration, the actual account types, the non-transferable positions, the held-direct funds nobody remembered. In a book of any size, the two disagree, and the disagreements are concentrated in exactly the complex households that score highest on friction risk. Reading statements at book scale is what document intelligence is for, and it produces the complexity score as a byproduct of the pre-validation you need to do anyway.
Score at the household, not the account. Roll accounts up to the household before scoring. A household with a clean brokerage account and one messy trust is a messy household, because the client experiences the slowest-moving piece.
Date the scoring. Do it close to the move. A score built four months before a resignation will be stale on the factors that matter most.
Using the score during the window
The score is a triage tool for a bounded period, not a permanent client segmentation. Three ways to use it once you are in the window.
Sequence the outreach. Highest risk-adjusted value first, on day one, personally. Not the largest accounts first. The largest low-risk household will likely still be there in week three. The medium-sized high-risk one will not.
Route the operations queue by the same ranking. If ten households can be pre-validated and submitted first, they should be the ones where operational delay costs the most, not the ones that happen to be easiest. This is the coordination that usually does not happen, because the ops queue is ordered by convenience and the outreach list is ordered by relationship, and nobody reconciles them.
Re-score at day 30. Households that have not signed by then are showing you their real risk level, which may differ from your estimate. Move them up. Households that signed immediately can drop out of active management.
Firms running many transitions at once, particularly transition consultants managing books for multiple advisors, benefit most from making this a standard artifact. The scoring is quick once it is a template, and it gives every deal a comparable risk picture rather than each advisor's personal sense of who might leave.
What the score cannot do
Be clear about the limits.
It does not predict individual outcomes. It ranks. A household scoring 15 may stay and one scoring 3 may leave for a reason nobody could see. The value is in allocating scarce attention, not in forecasting.
It does not substitute for the compliance boundaries on what you can say and when. What you are permitted to communicate to clients before and after resignation is governed by your agreements and your Protocol status, and no retention plan survives a solicitation problem.
It does not fix a slow repaper. If the operational track takes 90 days, no amount of outreach sequencing holds the book. The score identifies which households cannot tolerate delay, but the fix for delay is the transfer process itself, not the phone calls. Firms that have compressed the repapering timeline see the retention benefit directly, which is what the advisor transition platform is built around and what the advisor transition case study documents on a real book.
Context on why this matters more each year: Cerulli projects a large share of advisory assets changing hands over the coming decade as advisors retire and move, and its research on advisory asset transition frames the scale of movement the industry is absorbing. Clients are also better informed about their options than they used to be, with the SEC's investor guidance on working with investment professionals among the resources they now consult when a relationship changes. Neither fact changes the mechanics. Both raise the cost of being unprepared for the specific households most likely to reconsider.
Frequently Asked Questions
Which factor best predicts whether a client follows an advisor to a new firm?
Relationship ownership. Households the advisor personally originated follow at materially higher rates than households inherited from a retiring colleague or sourced by the firm. Advisors consistently overestimate their ownership of inherited relationships, which is why the scoring rubric asks whether you can name the client's family and their reason for investing rather than how long the account has been on your book.
Should outreach during a transition be sequenced by account size?
No. Size-ranked outreach systematically under-serves medium-sized, high-risk households, which is where surprise losses concentrate. Sequence by risk-adjusted value, meaning the product of household value and estimated risk, so a mid-sized fragile relationship gets attention before a large stable one that will still be there in week three.
How does repapering speed affect client retention?
Directly. Operational friction erodes a client's confidence in a decision they have already made. Each additional week of unsigned forms, confusing custodian mail, and positions in limbo gives the former firm's retention effort another opening. Households with complex or non-ACATS holdings carry both the highest friction and often the highest value, which is why they need expedited operational handling rather than standard queue position.
When should the attrition scoring be done?
Close to the move, as part of pre-move preparation alongside the account inventory and data reconciliation. Scoring done months in advance goes stale on the factors that change fastest, particularly life events and engagement. The complexity factor should be built from actual custodial statements rather than CRM records, since the two commonly disagree in exactly the households that score highest on risk.
Is it reasonable to decide not to pursue certain households?
Yes, and deciding deliberately is better than deciding by accident. Outreach capacity in the transition window is fixed. Households that are simultaneously high risk, low value, and operationally complex can consume the hours that would have retained a top relationship. Making that call in advance, in writing, is what prevents capacity from being allocated by whoever calls first.
