Most succession decks open with the same line: 10,000 Americans turn 65 every day. That number came from a Pew estimate built in 2010, and it now understates the wave. The Alliance for Lifetime Income puts the figure at 11,400 a day, with 4.18 million Americans reaching 65 in 2025 and the Peak 65 zone running from 2024 through 2027 (Alliance for Lifetime Income, 2025).
That correction matters. If your planning assumptions are 16 years old, so is your sense of urgency.
The bigger problem is what the headcount framing hides. Succession planning for the retirement wave is usually run as a coverage exercise: count the seats, name the successors, mark the plan complete. The people leaving are not seats. They hold the undocumented judgment that keeps critical processes running, and they know which vendor delivers in a crisis and which control matters in an audit. You can fill the role and still lose the capability.
This piece reframes retirement risk as capability risk: what the demographic data says, why the generation behind is constrained in availability rather than ambition, how to map exposure without asking anyone about their plans, and how to run knowledge transfer that survives the handover.
The demographic picture is not a cliff, and that is the harder problem
A cliff would be easy. You would see it coming and plan around a date. What the data describes is more diffuse and more dangerous.
At 11,400 a day across 2024 to 2027, Peak 65 is not a single quarter of departures. It is a four-year window in which the largest cohort of experienced workers in the American economy passes the traditional retirement threshold. Some leave immediately. Many stay years longer. Exposure spreads across the whole window, so it never announces itself.
Now add the projection data, because this is where most commentary goes wrong. The US Bureau of Labor Statistics Employment Projections for 2024 to 2034 show labor force participation among people aged 55 and over falling from 38.4% in 2024 to 36.9% in 2034 (BLS, 2025). Read alone, that looks like a straightforward exit story.
It isn’t. Over the same period the 55-and-over labor force grows from 38.806 million to 40.876 million, and the 75-and-over labor force rises from 2.131 million to 3.618 million, an increase of roughly 70% (BLS, 2025).
What falling rates plus rising numbers actually means
Both are true because the population is aging faster than the participation rate is falling. A smaller percentage of a larger group is still a bigger group. Three consequences follow.
First, you cannot assume mass simultaneous exit. Older workers are staying, and the number working past 75 is projected to keep climbing. Organizations that panic-plan for a cliff will over-hire and under-transfer.
Second, you cannot assume stability either. A falling participation rate means the probability of any given older worker leaving in a year is rising. Aggregate headcount holds up while individual predictability degrades. The org chart looks fine right up to the moment a specific person goes.
Third, tenure concentration deepens. If people stay longer, the knowledge held by one individual grows year over year and the eventual loss is larger. Long tenure is not a hedge against retirement risk. It is the mechanism that creates it.
Why succession planning for the retirement wave cannot be solved by hiring
The replacement math is unforgiving. The BLS projects total employment growth of just 5.2 million jobs from 2024 to 2034, reaching 175.2 million, an increase of 3.1% against 13.0% growth over the previous decade (BLS, 2025).
That is not a booming external market you can draw from. It describes a slow-growing labor pool in which the experienced segment ages out and the replacement inflow is thin. When external supply is constrained, internal transfer becomes the primary continuity mechanism. Whatever you do not transfer, you will try to buy in a market that does not have it.
Aging at the top, and the timetable boards set for themselves
The wave reaches the executive suite and the boardroom too. Spencer Stuart’s analysis of 2025 S&P 1500 CEO transitions puts the average age of departing CEOs at 61.8 (Spencer Stuart, 2025). The Conference Board, with Semler Brossy and ESGAUGE, reports the average sitting CEO age holding steady at around 58 across both the S&P 500 and the Russell 3000 from 2020 to 2025 (The Conference Board, 2025). A meaningful share of sitting chief executives are inside a four-year window of the typical departure age.
The board picture runs older still. The 2025 Spencer Stuart US Board Index reports an average independent director age of 63.6, and finds that 64% of S&P 500 boards now set mandatory retirement at 75 or older, up from 34% in 2015 (Spencer Stuart, 2025).
That last figure deserves attention. Over a decade in which boards pressed management to accelerate succession and deepen benches, they roughly doubled the share of themselves permitted to serve into their late seventies. This is an observation, not an accusation. Directors carry institutional memory that is hard to replace, which is exactly the argument a 40-year plant manager would make about her own role. It holds at every level, and an organization that accepts it in the boardroom while rejecting it on the shop floor has a problem.
The generation behind is deferring, not declining
The second half of the story is who inherits the work, and this is where commentary gets moralistic. The data does not support it. The Deloitte 2026 Gen Z and Millennial Survey, fielded to 22,595 respondents across 44 countries between November 2025 and January 2026, found that only 6% of Gen Zs and millennials name reaching a leadership position as their primary career goal (Deloitte, 2026). Taken alone, that reads as a generation walking away from management.
The same survey found that 76% of Gen Zs and 67% of millennials are interested in senior leadership at some point. The honest reading is deferral, not rejection. They want the role. They do not want it now, on current terms. The barriers they name are specific: stress and burnout (50% of Gen Z, 49% of millennials), too much responsibility (50% and 48%), and work-life balance (41% and 46%). Those describe the job as designed.
DDI’s research on younger leaders stepping back finds Gen Z is 1.7X more likely than other generations to step away from leadership roles to protect wellbeing, and that only 30% of leaders feel they have enough time to perform their duties effectively (DDI Global Leadership Forecast 2025). The same forecast reports that 71% of leaders experience increased stress.
Wider spans mean slower successor formation
Underneath it sits a structural change. Gallup’s span of control research, published in January 2026 and drawing on 16,442 managers plus a meta-analysis of 312 studies covering nearly 900,000 employees, tracks average direct reports per US manager rising from 8.2 in 2013 to 10.9 in 2024 to 12.1 in 2025, close to a 50% increase (Gallup, 2026). Gallup also identifies a workload threshold: managers spending 40% or less of their time on non-managerial work sustain 37% team engagement regardless of team size, while above that threshold engagement decays as teams grow, falling to 32% at 25 or more reports.
The succession consequence is clear. The person you are counting on to develop a successor manages half again as many people as their predecessor did, and is more likely to be over the threshold where the managerial part of the job gets squeezed out. Teaching goes first. Slower successor formation is not a motivation failure. It is arithmetic.
Skill churn compounds the loss
Retirement risk would be manageable if the job being handed over stayed still. It doesn’t. The World Economic Forum Future of Jobs Report 2025 projects that 39% of workers’ core skills will change by 2030, and finds leadership and social influence up 22 percentage points in importance since 2023, the largest increase after analytical thinking (WEF, 2025).
So the successor absorbs the departing expert’s judgment while two in five of the skills that job requires shift under them. Retirement risk and skill churn are not separate lines on the register. They multiply. The knowledge that transfers most cleanly, documented procedure, is also the most likely to go stale inside the transfer window. The knowledge that holds its value, judgment about ambiguity and people, is the hardest to move.
How to map retirement risk without asking anyone when they plan to retire
This is where organizations freeze or overstep.
US employment law restricts age-based decisions, and building plans on assumptions about who is “probably about to retire” creates real exposure as well as a poor employee experience. Pressuring people to disclose retirement intentions, or quietly withholding development from those you have written off, is a legal risk and a fast way to lose the people you need to stay long enough to teach. Confirm your approach with employment counsel first. This article is not legal advice.
The workable answer moves the unit of analysis from the person to the role. You are not forecasting when one individual will retire. You are assessing how exposed a critical role is to a continuity event, and retirement is one of several, alongside resignation, illness, promotion and recruitment by a competitor. Framed that way, the analysis applies the same standard to everyone.
Use structural signals, not individual assumptions
Three categories of data give a defensible picture without guessing at intent.
- Eligibility windows and benefit milestones. Plan documents define when eligibility opens, and defined benefit plans, retiree medical bridges and service-based accrual steps create further inflection points. The share of a role’s population that has crossed or is approaching those thresholds is a fact about the role’s workforce, not a judgment about an individual. Treat it as a population statistic reported in aggregate.
- Years of service and tenure concentration. This does more work than people realize, and it is not a proxy for age. A role where three people hold an average of 22 years of service and nobody else has more than three is fragile regardless of anyone’s birthday. It would be fragile if all three were 40. Tenure concentration measures how much operating knowledge sits with how few people.
- Role criticality. Independent of who holds it, how much damage does a 90-day vacancy cause? Weigh revenue or production continuity, regulatory and safety obligations, personally held customer relationships, single points of failure, and how long external replacement takes in the market described above.
Set thresholds so aggregate reporting never identifies individuals. Where you want individual input, make it voluntary and universal: a career-intent conversation offered to everyone in a critical role is legitimate development practice. An inquiry aimed only at older employees is not.
A risk score you can build this quarter
Scoring turns debate into a queue. Use four dimensions, each rated 1 to 5, with definitions written down so ratings compare across units.
- Criticality. Operational, financial, regulatory and relationship damage from a 90-day vacancy. 1 is absorbable, 5 is a stoppage or a reportable event.
- Continuity exposure. Concentration past eligibility or benefit milestones, combined with tenure concentration and role headcount. A single incumbent with 25 years of service scores 5.
- Readiness gap. The inverse of bench strength. 1 means a ready-now successor plus a developing second. 5 means nobody identified.
- Knowledge tacitness. How much of the role’s value sits in undocumented judgment, relationships and pattern recognition rather than written process. 5 means the work lives in one person’s head.
The composite equals criticality multiplied by the average of the other three, giving a range of 1 to 25. Bands: 1 to 8 monitor annually, 9 to 15 plan this year, 16 and above act this quarter with a named owner and a funded transfer window.
Two rules keep the model honest. Criticality multiplies rather than adds, so a low-criticality role with a terrible bench never outranks a critical role with a thin one. And someone other than the incumbent scores tacitness, because incumbents underestimate how much of their work is undocumented. Ask whoever covers the role during vacations. They know what breaks.
Run the scoring across your top 40 to 60 roles, not the whole population. You get a usable queue in weeks and a defensible rationale for where the money goes.
Knowledge transfer that actually transfers
Once the queue exists, the work is transfer. Most programs fail here, treating documentation as the deliverable.
Separate documented process from tacit judgment
Documented process is what the role does: the sequence, the systems, the approvals, the calendar. It is what most transfer efforts produce, and it is the easier half. A capable successor learns it from a well-written runbook.
Tacit judgment is what the role decides: which exceptions are safe to grant, which supplier claim is credible, which early signal in a monthly number means something, who to call when the formal escalation path is too slow. It is pattern recognition built over years, and the holder usually cannot articulate it on demand, because to them it feels like common sense rather than knowledge.
That changes what you ask for. Do not ask a departing expert to write down everything they know. Ask them to work through live decisions with a successor present, and capture the reasoning as it happens.
Four practices that move judgment
- Structured shadowing. Unstructured shadowing produces a successor who has watched a lot and learned little. Structure it around decision types rather than calendar time: identify the six to ten recurring decisions that carry the most risk, and make sure the successor sees each several times, with the incumbent narrating why, not just what. Then reverse it. The successor decides, the incumbent reviews. Reversal verifies the transfer.
- Overlap periods. A genuine overlap of eight to twelve weeks in a high-scoring role is the most effective single intervention, because the successor makes real decisions while the safety net is still in the building. Fund overlap by risk score, not seniority.
- Decision journals. A short standing entry per significant decision: what was decided, the alternatives, what evidence mattered, what the decision-maker expected, and what would have changed their mind. Ten minutes a week. Journals outlast the individual and capture reasoning rather than steps. They earn their keep long before anyone retires, which makes them sustainable.
- Communities of practice. Where a role exists in several locations or business units, a standing cross-site forum turns single-point expertise into distributed expertise. The controller at one site learns how three peers handle the same edge case. This lowers tacitness across a whole role family and is the cheapest structural fix available.
What to do when overlap is not funded
Often it isn’t. Finance approves the backfill on the departure date and not a day earlier. Three moves work without incremental headcount.
Start transfer 18 to 24 months out rather than at notice: the successor takes real accountability for a slice of the role now, while the incumbent is still there. Rotate coverage deliberately, so vacation and leave periods become planned transfer tests rather than survival exercises, then debrief what broke. And convert part of the role rather than the person: move the two or three highest-tacitness responsibilities to a shared model with a documented decision framework, so fragility drops even if the individual stays. Each costs less than the overlap you did not get.
Phased retirement, alumni networks and returnships as capability retention
These get dismissed as headcount tricks or favors to long-serving employees. Scored against continuity risk, they are transfer instruments.
Phased retirement, where an experienced employee reduces hours while staying engaged, buys transfer time you cannot otherwise purchase, and the BLS projection that the 75-and-over labor force will grow by roughly 70% by 2034 says the appetite exists. The design question is what the reduced-hours role is for. Keeping someone in the same job at 60% capacity wastes the opportunity. Reassigning them to coaching, decision review and documentation of the highest-tacitness work converts remaining time into transferred capability.
Alumni and returnship arrangements do similar work after the fact. A retired expert on contract for a specific transition or regulatory cycle is often the cheapest insurance against a transfer that did not complete. Set these up in advance, while the relationship is warm, with terms agreed and compliance questions resolved. Reaching someone 14 months later works less often.
Two cautions. Offer these on consistent, role-based criteria rather than case by case, for fairness and for the legal reasons above. And keep them purpose-bound, because a phased arrangement with no defined transfer outcome becomes a permanent one, delaying the succession it was meant to enable.
Put retirement risk on the same dashboard as readiness
None of this survives a busy organization unless it appears in the review that already happens, and the readiness baseline is not encouraging. DDI’s Global Leadership Forecast 2025, covering 10,796 leaders and 2,185 HR professionals across more than 50 countries, reports that internal candidates can immediately fill only 49% of critical positions, and that only 20% of HR leaders say they have leaders ready for their most critical roles (DDI, 2025). The upside is measurable: organizations with strong benches are 2.9X more likely to fill leadership roles internally and 2.8X more likely to outperform financially.
Bench strength and retirement exposure usually sit in different places, owned by different people, on different cycles. That separation lets a role look covered on the succession slide while carrying a single 27-year incumbent, no funded overlap, and a successor who has never made the hard call alone.
Three changes fix it. Show continuity exposure and readiness on the same view, so no role is marked green on one dimension while red on the other goes unmentioned. Track transfer progress as a status rather than successor names: the percentage of critical decisions a successor has made under supervision tells you more than a name in a box. And review the top of the queue quarterly with the seriousness of a financial forecast, with named owners and dates.
That is a data problem before it is a technology problem, though it gets easier when one record holds both halves. BullseyeEngagement’s Succession Planning and AI Advisor capabilities help leaders bring competencies, performance history, readiness data, successor coverage, development needs, and exposure risk into a more structured view. On the Workforce Planning side, scenario modeling by business unit, division, or location helps leaders understand where future workforce risk may emerge and where action is needed.
For the metrics layer, our guide to succession planning metrics and KPIs covers what is worth reporting to a board, and our piece on integrating demographic trends into workforce planning goes deeper on the projections.
Frequently asked questions
1. How many Americans turn 65 every day in 2026?
About 11,400 Americans turn 65 each day during the Peak 65 zone, which spans 2024 through 2027, with 4.18 million reaching 65 in 2025 (Alliance for Lifetime Income, 2025). The widely quoted 10,000 a day comes from a 2010 Pew estimate and understates the wave.
2. Can we legally plan succession around employees who are near retirement age?
Plan around roles, not individuals. Use role criticality, tenure concentration and aggregate eligibility data rather than assumptions about anyone’s intentions, and never withhold development based on perceived retirement proximity. Employment law restricts age-based decisions, so confirm your approach with employment counsel first.
3. How do you score retirement risk for a critical role?
Rate four dimensions from 1 to 5: role criticality, continuity exposure from tenure and eligibility concentration, readiness gap, and knowledge tacitness. Multiply criticality by the average of the other three. Scores of 16 or above need action this quarter with a named owner.
4. Is the older workforce shrinking or growing?
Both, depending on the measure. BLS projects participation among people 55 and over falling from 38.4% to 36.9% between 2024 and 2034, while the 55-plus labor force grows from 38.8 million to 40.9 million and the 75-plus group grows by roughly 70% (BLS, 2025). Rates fall while numbers rise.
5. Do younger employees want leadership roles?
Mostly yes, later. Only 6% of Gen Zs and millennials name leadership as their primary career goal, but 76% of Gen Zs and 67% of millennials are interested in senior leadership at some point (Deloitte, 2026). The barriers they cite are stress, excessive responsibility and work-life balance, which are design issues.
