An employee hands in their notice. HR schedules the exit interview, IT revokes system access, the manager posts the job opening. Every box on the offboarding checklist gets ticked. And yet, three months later, someone on the team is stuck trying to figure out why a client account was configured a certain way, or which vendor relationship actually mattered, or what the “quick fix” was for a recurring production issue. The person who knew is gone, and the checklist never asked them.
This is the part of offboarding that most organizations don’t design for. Laptops get returned. Badges get deactivated. Knowledge, the kind that lives in someone’s judgment rather than a document, rarely gets captured at all.
The gap is bigger than most leaders assume
Research from Panopto found that on average, 42 percent of the skills and expertise required to perform a given role are known only by the person in that role. Not documented anywhere. Not shared with a backup. Held in one person’s head, and gone the day they leave.
That figure lines up with what APQC found when it studied the wave of retirements now moving through the workforce. In APQC’s research, 41 percent of organizations admitted they rarely or never attempt to collect know-how from employees before they retire. Where an attempt does happen, it is usually a single exit interview in the employee’s final week, which is a reasonable gesture and a poor substitute for actually transferring thirty years of judgment.
The scale of the exposure is not abstract. IDC estimated the cost of knowledge loss at a company with 1,000 employees and 7 percent attrition at roughly $300,000 per week, measured through the time employees spend searching for information that already exists somewhere, or recreating it because they can’t find it. At the enterprise end, the number gets larger still: Fortune 500 companies are estimated to lose at least $31.5 billion a year from failures to share knowledge internally.
None of this is because organizations don’t care. It’s because offboarding was designed as an HR and IT process, not a knowledge process, and nobody owns the gap between the two.
Why the standard exit interview doesn’t solve it
The exit interview exists to answer a different question than the one that matters here. It’s built to capture why someone is leaving, what they thought of the culture, whether they’d recommend the company to a friend. Useful information, but not knowledge transfer. It asks the departing employee to reflect on their experience. It does not ask them to hand off their judgment.
Three things make the standard exit interview structurally unsuited to the job:
- Timing. It happens in the final days or the final week, when there’s no runway left to actually walk a successor through anything nuanced.
- Format. It’s a single conversation, usually with HR rather than a subject-matter peer, which means the questions rarely go deep enough to surface tacit judgment, only formal responsibilities.
- Ownership. Nobody is accountable for what happens to the information afterward. If it isn’t immediately actionable, it sits in a notes file that nobody revisits.
Tacit knowledge, the unwritten, experience-based judgment someone builds over years, doesn’t surface under those conditions. It surfaces through structured elicitation: specific questions about specific decisions, asked early enough that there’s still time to record, verify, and transfer the answers to someone who will actually use them.
What structured knowledge transfer looks like in practice
APQC’s work on structured knowledge transfer offers a useful distinction here: the goal is not to document everything a departing employee knows. That’s neither possible nor useful. The goal is to identify what is critical, tacit, and not held anywhere else, and transfer that specifically.
A few practices consistently separate organizations that retain knowledge from those that don’t:
1. Start the process at notice, not at departure
The moment someone gives notice (or, for planned transitions like retirements, well before that) is the moment a structured handoff should begin, not the final week. Even two to four weeks of overlap changes what’s possible: enough time for a successor to shadow real decisions, not just read a summary of them.
2. Ask about decisions, not duties
A job description tells you what someone was responsible for. It doesn’t tell you how they decided things when the standard process didn’t apply. The higher-value questions sound like: “What’s a judgment call you make regularly that isn’t written down anywhere?” or “What would a new person get wrong in their first six months that you’d catch immediately?” These questions surface the 42 percent that documentation misses.
3. Pair capture with a named recipient
Knowledge captured with no destination tends to sit unused. Structured transfer works best when there is a specific successor, team, or community of practice designated to receive it, ideally someone who can ask follow-up questions while the departing employee is still reachable.
4. Treat relationship knowledge as seriously as technical knowledge
Client history, vendor context, and internal political dynamics rarely make it into any handoff document, yet they’re often what determines whether a transition is smooth or costly. A senior account manager’s departure, for instance, frequently puts a client relationship at risk not because the successor lacks technical competence, but because they lack the relationship context the predecessor never wrote down.
The cultural piece leaders tend to skip
None of the above works if departing employees have reason to hold information back, and in many organizations, they do. When knowledge sharing is never rewarded and expertise is the main source of someone’s job security, there is a quiet incentive to keep some of it close. This is not usually a character issue. It’s usually a structural one, created by how the organization has treated knowledge sharing up to that point.
Organizations that get more out of departure knowledge transfer tend to frame it as a form of recognition rather than an extraction exercise, positioning the process as capturing someone’s expertise for its own value, not as a final obligation on the way out the door. That framing shift costs nothing and changes how forthcoming people are willing to be in the conversations that matter most.
The takeaway for KM and people leaders
Offboarding checklists were built to protect systems and assets. They were never built to protect judgment, and judgment is what actually walks out the door. Closing that gap doesn’t require new software or a large program. It requires treating the weeks before a departure as a knowledge transfer window, asking better questions during that window, and giving the answers a specific home. Organizations that build even a lightweight version of this into their offboarding process stop losing the knowledge that never had a chance to be written down in the first place.