Knowledge Transfer in Offboarding: Six Objections to the Last Four Weeks

Handover meeting, checklist, remaining leave: why knowledge still walks out with the person, and how to capture it in two weeks.
Anna Müller
instagram iconlinkedin icon
Illustration: a person walks toward the door, their knowledge stays behind as a trail and is captured

Between a resignation and the last working day there are usually a few weeks, and in that time it is decided what of a person's knowledge stays with the organisation. Every attempt to use those weeks meets the same six sentences, and each of them has a true core.

"The manager will cover it in the handover meeting"

Almost every organisation has a handover meeting, and it is better than nothing. It has one built-in flaw: the manager asks about what the manager knows. Current projects, open deadlines, contacts, where which file lives.

What actually made the person irreplaceable is usually something the manager does not know, otherwise it would not be a problem. The exception for the key account that was never put in writing. The trick that gets the line running again after a stoppage. The colleague in accounts you call when an invoice is stuck. None of this comes up in a conversation about responsibilities. It comes up when someone asks about specific cases, and keeps asking.

In nearly three out of four companies, losing one particular person would noticeably hurt operations (SHRM, 2023). For exactly those people, a chat in the doorway is not enough.

"Once they've resigned, they've checked out"

Sometimes that is true. Often it is not. More often it is an assumption that confirms itself. Signal to the leaver that from now on they are only being processed, and that is exactly what you get back.

Experience from knowledge capture projects points the other way. Many people who leave want their work not to disappear with them. That is not idealism, it is professional pride: you built something, and it is unpleasant to imagine that in three months nobody can do it any more. What really slows participation is rarely a lack of will. It is effort. Ask someone to write a manual in their last weeks and no manual gets written.

The difference lies in the format. A conversation in which someone asks targeted follow-up questions costs the person half an hour and feels like recognition. An empty document titled "Handover" feels like a chore and is treated like one.

"The successor isn't in place yet, so there's nothing to hand over"

This objection does the most damage, because it sounds logical. In most minds, handover means person A explains to person B how it works. If person B is missing, you wait.

But the waiting has a fixed end, and that is the last working day. Barely one company in five has a formal succession plan (SHRM). Everywhere else the successor is often not there when the leaver goes, and then the knowledge goes to nobody.

The answer is to separate handover from succession. The knowledge is captured now, not for one person, but so that it stays verified and retrievable. The successor arriving in three months finds it waiting. So does the one after that. That is the real difference between a handover and a safeguard, and the article about dissolving single points of knowledge goes deeper into it.

"We have an offboarding checklist"

The checklist is good. It makes sure the laptop comes back, access is revoked and the reference letter is ready on time. HR built it for HR processes, and for that it works.

Knowledge usually appears on it as a single line: "Handover completed". A tick that says nothing about what was handed over or whether anyone can find it again.

If you want to extend the checklist seriously, two questions asked in the first week after notice are enough. On which matters is this person regularly asked even though they are not formally responsible? And which of those come up less than once a quarter? The second question matters more. What happens every week, the successor learns on the job. What happens once a year catches them unprepared.

"Remaining leave eats the time anyway"

True.

With remaining holiday and overtime, a four-week notice period often shrinks to two, sometimes less. And day-to-day business has to keep running in that time.

Which is exactly why the question is not whether to use those weeks, but how little of the person's time it takes. A project that costs half a day per topic fails on arithmetic. An AI-guided interview needs about 30 minutes, and the system asks the follow-up questions itself, wherever an answer stays vague. Three to five of those conversations fit into two weeks. How such an interview works is described in the post on AI expert interviews.

In Omnora Intelligence Management, the result is not a transcript but verified knowledge with a named owner, a validity date and a review status. It can be searched, turned into training and used by AI applications. That is the part no checklist can deliver.

"If need be, we'll just give them a call"

This is the most honest objection, because it describes what actually happens. After the departure, the former colleague gets a call, and usually she helps.

That works for a while. Then she changes her number, has no time at the new company, or simply cannot remember the case from two years ago. And when it gets serious, she comes back as a consultant, which typically costs two to three times her former day rate (HBR). You then pay for knowledge the organisation already owned once.

That leaves a question every manager can ask today, without anyone having resigned: who would you call if this person left tomorrow?

If the only name that comes to mind is hers, four weeks are already too short.

Similar posts