# Retirement Cliffs And Why Firms Keep Walking Off Them **By Dan Cumberland** · Published August 15, 2026 · Categories: AI Strategy > A retirement cliff is not a metaphor. When a senior project manager with 26 years of experience walks out the door, the firm loses something that doesn't... A retirement cliff is not a metaphor\. When a senior project manager with 26 years of experience walks out the door, the firm loses something that doesn't appear on any balance sheet: which subcontractor won't inflate change orders under pressure, why a particular hillside site failed three winters ago, the design decision logic from a career's worth of projects that didn't go as planned\. That knowledge doesn't transfer automatically\. It retires on the same Friday she does\. And it doesn't come back\. What makes this different from ordinary turnover is the compression\. Normal attrition gives firms time and overlap\. The retirement cliff doesn't\. Gartner research[3](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-3) estimates that 70 to 80 percent of enterprise knowledge is tacit— it has never been written down in any retrievable form\. When the person exits, the knowledge goes with them\. That scenario is playing out across the AEC industry right now— at a scale most firm leaders haven't fully processed\. This article names the five organizational patterns that cause firms to know this is coming and still don't act— and what the firms that handle it well do differently\. ## The Scale of the Problem Is Bigger Than Most Firms Have Processed More than 41 percent of the AEC workforce is projected to retire by 2031, per NCCER data[4](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-4)\. In raw terms: approximately 10,000 Americans turn 65 every day through 2030[7](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-7)— the largest simultaneous knowledge exit in modern workforce history\. The firm\-level math is more immediate\. Replacing a departing knowledge worker costs 50 to 200 percent of their annual salary, and new hires take 8 to 12 months to reach full productivity[3](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-3)\. Multiply that by a senior PM, a lead estimator, and a principal— all retiring in the same fiscal year— and the disruption compounds\. Deloitte estimates $6\.9 to $9\.6 trillion in lost U\.S\. economic output from boomer retirements over the next four years[2](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-2)\. The "Great Retirement"— APQC's term for the concentrated boomer workforce exit between 2025 and 2030— is creating a timeline that doesn't leave room for gradual adjustment\. By 2030, an estimated 61 million baby boomers will exit the workforce[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)\. Construction alone needs 499,000 new workers in 2026[4](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-4)\. But here's the part most analysis misses\. AEC firms face a compound problem: ```html-table
FactorData Point
Senior workforce exiting41% of AEC workforce by 2031 (NCCER)
Average AEC firm tenureFell from 7.1 years (2012) to 4.9 years
Net effectLess knowledge-transfer overlap than firms expect
``` Boomers are leaving faster\. And average tenure at architecture and engineering firms has already dropped from 7\.1 years to 4\.9 years[4](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-4)\. The knowledge transfer window is shrinking from both ends simultaneously\. The question isn't whether the wave is real\. The more interesting question is what, exactly, firms are losing— and whether it can be recovered\. ## The Real Loss Isn't the Job Description— It's Everything That's Not Written Down Gartner estimates that 70 to 80 percent of enterprise knowledge is tacit— it exists in people's heads and has never been written down in any retrievable form[3](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-3)\. For AEC firms, that means the "how" behind a senior engineer's decisions, their institutional relationships with key subcontractors, and the pattern recognition from hundreds of site conditions all exist in exactly one place: the person\. Three categories of knowledge walk out that documentation alone can't replace: - **Judgment**— the accumulated reading of situations that experience builds and org charts don't capture - **Relationships**— the trust and rapport with clients, subcontractors, and regulators that was built over years of showing up - **Pattern recognition**— the ability to see what's about to go wrong before it does, often without being able to explain why How rarely this knowledge actually transfers is striking\. In a survey of 1,500 baby boomers[6](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-6), 57 percent had shared half or less of the knowledge needed to perform their job responsibilities before retirement\. Twenty\-one percent had shared none\. The consequences show up in adjacent industries too\. Banks that let experienced COBOL programmers retire without knowledge transfer ended up rehiring those same people as contractors at premium rates— a "perpetuated dependency cycle" that costs more and recovers less than capture would have\. Boeing's safety challenges have been partially attributed to the loss of experienced engineering judgment through outsourcing and layoffs— one factor among several[6](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-6), but a documented one\. And that's the loss\. The harder question— the one I haven't seen clearly answered anywhere— is why firms that understand this keep failing to act on it\. ## Five Reasons Firms Keep Walking Off the Same Cliff Ninety\-two percent of organizations fail to consistently capture knowledge from soon\-to\-be retirees[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)\. What's striking: 85 percent of C\-suite leaders already view the knowledge exodus as a moderate to mission\-critical threat[2](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-2)\. This isn't ignorance\. It's a pattern— and it has a name\. ### 1\. The Awareness\-Action Gap Most firms know this is coming\. Forty\-one percent still rarely or never attempt to collect know\-how from retiring employees[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)— even when the departure is scheduled and visible months in advance\. Knowing about a cliff and changing course before you reach it are two different decisions\. Most organizations get stuck between them\. Crossing that chasm from awareness to action is the real work\. ### 2\. The "When We Have Time" Trap Fifty\-two percent of organizations say day\-to\-day deadlines leave little room for knowledge\-sharing sessions[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)\. That's the honest answer\. It's also a trap\. Those deadlines don't disappear\. And the project schedule doesn't clear out to accommodate knowledge transfer\. Firms that wait for bandwidth never find it— and the senior employee retires right on schedule\. The fix isn't "making time\." It's embedding knowledge capture in the work itself, not treating it as a separate project\. ### 3\. The Indispensability Paradox A dynamic worth naming: in some firms, senior experts hold their expertise closely— not out of malice, but because knowledge is the source of their influence\. Zweig Group has documented how AEC firm owners maintain indispensability by retaining sole decision\-making authority rather than delegating[5](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-5)\. As Zweig puts it: "If your firm cannot survive without you, it isn't really a firm— it's a job with overhead\."[5](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-5) That pattern isn't limited to owners\. It shows up at every level where expertise is concentrated and organizational structures don't create incentives to share it\. ### 4\. The Tenure Double\-Bind Even when firms notice a retirement coming, the transfer window is shorter than expected\. This is the compound problem revisited as a failure pattern\. Baby boomers average more than 8 years at a single employer[2](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-2)— far longer than the current AEC workforce average\. Average AEC tenure has fallen from 7\.1 to 4\.9 years[4](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-4)\. The junior employees who are supposed to receive knowledge have less runway with the senior employee than their predecessors did\. Both ends of the transfer window are compressing at once\. ### 5\. The AI Readiness Accelerant Here's what makes this urgent beyond the immediate operational loss\. Firms that don't capture institutional knowledge before retirement will find their AI initiatives running on empty— the tools work, but the context they need isn't there\. Gartner projects that 40 percent of agentic AI projects— AI systems designed to take autonomous action— will be cancelled by 2027, citing insufficient organizational context as a key obstacle[3](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-3)\. That context is exactly what retires with senior employees\. Bluebeam's CEO put the AEC version of this plainly: the real obstacles to AI adoption are "not cost, but complexity, culture, and connection\."[4](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-4) Building a [knowledge\-sharing culture](/blog/building-ai-culture) is prerequisite infrastructure for any AI initiative\. The tools can't help you if the knowledge they need isn't captured\. This is why knowledge transfer isn't just an HR problem— it's a technology strategy problem\. The good news: these patterns are visible from the outside, which means they're avoidable\. ## What Firms That Handle This Well Do— and When They Start The firms that handle the retirement cliff well share one trait that shows up before anything else: they got curious before they got worried\. They don't wait for exit interviews\. Best\-performing organizations begin structured knowledge capture programs three to five years before projected retirement dates, per APQC research[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)— long enough for real, repeated knowledge transfer rather than a rushed documentation sprint in the final weeks\. Three things distinguish firms that do this well: - **Start early\.** Three to five years before the projected retirement date, not six months\. This is the single most consistent differentiator APQC found[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)\. Firms that wait until the departure is announced are already too late for real transfer— they get documentation sprints and rushed handoffs, not knowledge\. - **Embed in daily work\.** Knowledge capture that lives outside the normal workflow fails because it competes with billable time— the same constraint behind Pattern 2\. The firms that succeed make documentation part of how senior people do their actual work, a performance criterion rather than an afterthought\. - **Prioritize mission\-critical knowledge first\.** Deloitte's analysis[2](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-2) found that 20 percent of knowledge content resolves 80 percent of issues\. Don't try to capture everything\. Start with what would hurt most to lose in the next six months\. Once that knowledge exists in a structured, transferable form, AI can transmit it at scale\. A European telecom company that implemented knowledge management saw issues resolved on the first call improve by 37 percent, net promoter score— a measure of customer loyalty— rise by 30 points, and new hire ramp time cut in half[2](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-2)\. But that outcome requires the knowledge to exist first\. The sequence matters\. For a deeper look at the AI approach, [how the best AEC firms approach institutional knowledge capture and AI](/blog/institutional-knowledge-aec/) covers the five\-step roadmap in detail\. A solid [AI governance strategy for professional services firms](/blog/ai-governance-strategy) provides the policy layer that makes any knowledge capture program sustainable\. ## The Cliff Is Visible from Here Every firm in the AEC industry is watching the same retirement wave coming\. What varies is whether they've made knowledge capture part of how they work— or whether they're still planning to "get to it\." But most haven't\. The firms that will thrive through the next five years of boomer retirements are the ones treating institutional knowledge capture as infrastructure, not a project\. No matter the question, people are the answer\. The knowledge lives in the people\. The goal is to make sure it doesn't walk out the door with them\. The starting point is internal: audit which senior people hold knowledge that isn't documented anywhere\. The three\-to\-five year lead time APQC recommends starts now, not after the retirement is scheduled\. If you're ready to build that infrastructure, [AI implementation services](/services/ai-implementation) at Dan Cumberland Labs help AEC firms capture what's in senior experts' heads— and pair it with AI systems that transmit it at scale\. ## Frequently Asked Questions ### What is a retirement cliff in business? A retirement cliff is the rapid, concentrated loss of institutional knowledge when a large cohort of experienced employees exits the workforce simultaneously\. Unlike routine attrition— where knowledge transfers gradually through overlap— a retirement cliff compresses the loss into a window shorter than most organizations can absorb\. Baby boomers \(born 1946–1964\) reaching retirement age simultaneously is the current version of this event, which APQC has named "The Great Retirement\."[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1) ### How many baby boomers are retiring? Approximately 10,000 Americans turn 65 every day through 2030[7](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-7), with an estimated 61 million baby boomers projected to exit the workforce by decade's end[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1)\. In the AEC industry specifically, NCCER projects that more than 41 percent of the workforce will retire by 2031[4](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-4)\. ### What does institutional knowledge loss cost a firm? At the individual level, replacing a departing knowledge worker costs 50 to 200 percent of their annual salary, with new hires taking 8 to 12 months to reach full productivity[3](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-3)\. At the macro level, Deloitte projects $6\.9 to $9\.6 trillion in lost economic output in the U\.S\. over the next four years from boomer retirements[2](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-2)\. ### Why don't firms prepare for the retirement cliff? Three barriers dominate, per APQC's 2025 Great Retirement study[1](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-1): 52 percent of organizations say daily deadlines leave no room for knowledge\-sharing sessions; 45 percent cite resource constraints; 35 percent say their culture doesn't support knowledge sharing\. The problem is organizational, not technical— the tools exist\. The bottleneck is will and priority\. ### What is tacit knowledge and why does it matter? Tacit knowledge is expertise that lives in someone's head— judgment calls, pattern recognition, client relationship context, and the reasoning behind decisions— that has never been formally documented\. Gartner estimates 70 to 80 percent of enterprise knowledge is tacit[3](/blog/blog-retirement-cliffs-and-why-firms-keep-walking-off-them#ref-3)\. It's the most valuable category and the hardest to transfer, which is why the retirement cliff matters so much more than a simple headcount loss\. ## References 1. APQC, "The Great Retirement: Knowledge Loss, AI and the Workforce Shift" \(2025\)— [https://www\.apqc\.org/resource\-library/resource\-listing/great\-retirement\-knowledge\-loss\-ai\-and\-workforce\-shift](https://www.apqc.org/resource-library/resource-listing/great-retirement-knowledge-loss-ai-and-workforce-shift) 2. Deloitte Insights, "The $9 Trillion Knowledge Exodus: How Organizations Can Turn Baby Boomer Retirements Into a Competitive Advantage" \(2025\)— [https://www\.deloitte\.com/us/en/insights/topics/talent/knowledge\-management\-plan\.html](https://www.deloitte.com/us/en/insights/topics/talent/knowledge-management-plan.html) 3. Atlan, "Institutional Knowledge Loss: Causes, Costs, and Prevention" \(2024\) \(citing Gartner 2024 and SHRM 2023\)— [https://atlan\.com/know/data\-for\-ai/institutional\-knowledge\-loss/](https://atlan.com/know/data-for-ai/institutional-knowledge-loss/) 4. DeepHumanX, "The AEC Knowledge Cliff: Why Generic AI Makes the Problem Worse" \(2025\) \(citing NCCER, ABC, BLS, and Bluebeam\)— [https://deephumanx\.com/resources/aec\-knowledge\-cliff\-organizational\-intelligence\-construction](https://deephumanx.com/resources/aec-knowledge-cliff-organizational-intelligence-construction) 5. Zweig Group, "The Succession Mistake Too Many A/E Firm Owners Still Make" \(2025\)— [https://zweiggroup\.com/blogs/the\-zweig\-letter/succession\-planning\-aec\-firm\-owners](https://zweiggroup.com/blogs/the-zweig-letter/succession-planning-aec-firm-owners) 6. Welcome to the Jungle, "Retaining Critical Knowledge as Baby Boomers Retire" \(2024\)— [https://www\.welcometothejungle\.com/en/articles/retirement\-surge\-brain\-drain](https://www.welcometothejungle.com/en/articles/retirement-surge-brain-drain) 7. U\.S\. Census Bureau, "U\.S\. Workforce Is Aging, Especially in Some Firms" \(December 2025\)— [https://www\.census\.gov/library/stories/2025/12/older\-workers\.html](https://www.census.gov/library/stories/2025/12/older-workers.html) --- ## About the author **Dan Cumberland** — Founder, Dan Cumberland Labs Dan Cumberland helps engineering and construction firms see where they stand with AI and decide what to build first. He created Pacemark, the AI maturity model behind that work, from research on more than 300 companies. - Take the assessment: https://pacemark.ai/signal/assessment/?track=aec&utm_source=dcl-site&utm_medium=link&utm_campaign=pacemark-assessment - Book a call: https://book.dancumberland.com/ai-strategy --- Source: https://dancumberlandlabs.com/blog/retirement-cliffs-and-why-firms-keep-walking-off-them/