What is tacit knowledge transfer?

It is the process of passing on know-how that lives in a person's experience rather than in documents: how an expert reads a situation, which signals they trust, who they call and why. Because it is hard to write down, it is mostly transferred by working alongside the expert, doing real tasks with them nearby, and hearing the stories behind their decisions.

Is a handover document enough when an expert retires?

Usually not on its own. A handover document captures the parts of the work that can be written down, which are often the easier parts. The judgment calls, the exceptions and the relationships need practice and conversation. Write the document, but plan time for shadowing, guided practice and a task the successor leads while the expert watches.

How long does knowledge transfer between an expert and a successor take?

It depends on how much is at risk and how often the critical tasks occur. As a starting point, six sessions over three to six months leaves room for the mapping, the practice and the review; adjust to the role. If a key task only happens once a year, plan the transfer around that date rather than around the calendar.

What if the expert is leaving in a few weeks?

Compress and prioritize. List the critical tasks, pick the two or three that would hurt most if nobody could do them, and spend the remaining time on those. For each of the rest, write a short note and name another person in the organization who holds part of the knowledge.

Does the successor have anything to teach the expert?

Often, yes. A successor sees the work with fresh eyes and may know tools or methods the expert does not use. Asking them what they would change can improve both the way the work is done and the way the knowledge is captured, and it gives the successor practice in leading.

Tacit Knowledge Transfer: How Experts Pass On What Is Not Written Down

Tacit knowledge transfer is the deliberate passing on of the know-how an experienced person carries in their head and hands rather than in documents.

The aim is that a successor can do the work well after the expert has retired or moved role.

It covers what makes someone good at a job:

the judgment calls
the exceptions
the warning signs
the relationships

It happens mostly through watching, doing and talking, not through reading.

This guide is for both people in such a pair.

One is the expert who is passing on what they know. The other is the successor or learner who is taking it over.

It explains how to tell written knowledge from experience, and how to learn the part that is not written down.

It also shows how to capture judgment so it outlasts the two of you.

Finally, it covers how to check at the end whether the transfer actually worked.

What is the difference between written knowledge and deep expertise?

Ikujiro Nonaka and Hirotaka Takeuchi distinguish two kinds of knowledge.

Explicit knowledge can be put into words, documents and procedures. Tacit knowledge lives in people’s experience.

Dorothy Leonard and Walter Swap use the term “deep smarts” for experience-based expertise.

They argue that cultivating it and passing it on is a leadership task.

In practice, almost every critical task contains both.

Take a monthly financial close. The steps are in a checklist.

What is not in the checklist:

which account always throws an error in March
which business unit sends late numbers, and how to chase them without causing a fight
the feeling that a total “looks wrong” before any check has failed

A successor who reads the checklist knows the steps.

A successor who has worked three closes next to the expert knows the job.

This is why the first question in any transfer is not “what do you know?” but “where does each piece of this live?”

Why does a handover document rarely work on its own?

Nonaka and Takeuchi’s SECI model describes four ways knowledge moves:

socialization (tacit to tacit, by working alongside someone)
externalization (putting experience into words)
combination (joining written pieces into something new)
internalization (turning what was read into practice)

A handover document is pure externalization and combination.

It skips the two movements that need people in the same room.

The model is contested among researchers, particularly in how it treats tacit knowledge.

You do not need to settle that debate to use the model as a checklist.

If your plan only has a document in it, you are missing socialization and internalization.

Then the successor will meet the hard parts alone.

A practical sign that a handover is document-heavy: the file runs to dozens of pages.

Yet the successor still phones the expert in the first week with a question the file does not answer.

How do you map what is at risk?

Start with a short list of critical tasks: five to eight things that would hurt if nobody could do them next year.

Include informal ones.

Examples are the supplier who only answers the expert’s calls, or the problem colleagues always bring to the same desk.

Then, for each task:

Find what is written. Collect the procedures, checklists and files that exist.
Check it. Ask the expert to describe the last time they did the task, step by step, and compare with the document. Where they departed from it, ask why.
Sort it into three piles. Written and correct. Written but incomplete or out of date. Held only in experience.
Pick a method per pile. The successor reads and checks the first, and updates the second with the expert. The successor learns the third by practice and conversation.
Record a baseline. The successor rates themselves 1 to 5 on each task. The expert rates them privately. Compare.

The baseline is what makes the end of the transfer measurable.

Without it, both people tend to feel the transfer went well, and nobody can say how well.

How do you learn by watching and doing?

David Kolb describes learning as a cycle of four stages:

a concrete experience
reflective observation
abstract conceptualization (drawing a rule from it)
active experimentation (trying the rule)

Shadowing on its own gives only the first two.

A good session runs the whole cycle.

Choose a real task, timed to when it happens. A demonstration on a quiet day hides the very situations that need judgment.
The expert thinks aloud. What they look at, what they ignore, what they decide and why. Experts often go silent when they concentrate, because the steps have become automatic. If that happens, record a short commentary straight afterward.
The successor takes three-column notes: what the expert did, what they looked at, and why. The third column comes from questions at a natural pause.
Hand over one step. The successor does it with the expert beside them, saying their decisions out loud first.
Debrief and write rules of thumb. Within two days, the successor writes three rules in their own words. The expert corrects them.

Over several sessions, the share the successor does grows.

In the end they lead a whole task while the expert observes.

The expert only steps in for real risk to safety, customers or compliance.

How do you capture judgment calls so they outlast the pair?

Judgment is built from cases, so the best way to pass it on is through cases.

Ask the expert for three kinds of story:

a decision that went well, and why;
a near miss or mistake, and what they check now because of it;
a situation where the procedure did not apply.

After each story, the successor asks what the expert noticed first and what options they considered.

The successor also asks what made them choose, and what they would do differently today.

Then they write it in five short lines: situation, early signals, decision, reason, lesson.

Reading it back to the expert often shows that the most important signal is missing from the first draft.

Two more things belong in the capture.

The first is a “who to call” map for each task.

It shows who to call when it goes wrong, who must never be surprised, and whose informal agreement is needed before acting.

The second is a home for all of it that does not depend on either person’s account.

Ideally it is attached to the procedure itself as a “notes from experience” section, so the successor’s own successor finds it.

Stories often involve colleagues and customers by name.

Capture the lesson, leave out the gossip, and check what your organization allows to be written down.

What does the successor bring back?

Wendy Murphy’s research on reverse mentoring describes a junior employee mentoring a senior one.

It has two purposes.

One is sharing knowledge such as technology and a generational perspective. The other is developing the junior person as a leader.

A knowledge transfer pair can borrow the idea without running a formal program.

The successor sees the work with fresh eyes.

Ask them which habits look odd from outside and which manual step a tool could handle.

Ask how the captured notes could be easier to search.

Some of the expert’s methods exist for good reasons that are worth explaining; some are only habit.

Either way, the conversation shifts the relationship toward the moment the successor works alone.

For more on how these conversations work, see the guide to reverse mentoring.

What does this look like in practice?

Three illustrative examples, composites rather than real cases.

A maintenance engineer retiring from a production site.

At risk: restarting a line after an unplanned stop, and the informal relationship with the equipment vendor’s field technician.
Written: the restart procedure, partly out of date after a refit.
Experience only: the sounds and readings that mean “wait before the next step”.
Method: the successor updates the procedure with the engineer, then shadows two real restarts and leads a third. The engineer introduces the successor to the vendor technician in person.

A key account manager moving to a new role.

At risk: the renewal negotiation with the largest customer, due in five months.
Written: the contract, the account plan and the CRM history.
Experience only: who really decides on the customer side, which topics end a meeting badly, and what was promised informally three years ago.
Method: three stories about past renewals captured in the five-line format. A “who to call” map for the customer’s organization. The successor also leads one quarterly review while the account manager observes.

A specialist who maintains an old internal system.

At risk: the yearly data migration and the fixes for errors nobody else has seen.
Written: almost nothing.
Experience only: nearly everything.
Method: because the migration happens once a year, the transfer is planned around that date. The successor proposes a searchable log of past errors, which the specialist fills in. The two of them run a test migration together beforehand.

How should you use this with your mentor?

If you are in a structured mentoring program for knowledge transfer, a workable session plan follows the same order as this guide.

Kick-off. Agree the list of critical tasks and record the baseline ratings. Settle where shared notes will live.
Mapping. Sort each task into the three piles and choose a method for each.
Watching and doing. One real task, with thinking aloud, a step handed over and rules of thumb written afterward.
Judgment calls. Three stories, a “who to call” map and two introductions.
The other direction. The successor teaches or proposes something, then leads a task.
Review and close. Re-rate, compare with the baseline, and name another person for every remaining gap.

The successor should own the notes, because they will use them longest.

The expert should push the task list past the obvious.

That is because much of what they know no longer feels like knowledge to them.

Both should prepare for each meeting.

The guide on how to prepare for a mentoring meeting covers the basics.

Questions mentors ask is a useful source for the expert’s side of the debriefs.

How do you know the transfer worked?

Melenie Lankau and Terri Scandura describe two kinds of personal learning in mentoring.

They are relational job learning and personal skill development.

Both matter here.

Ask whether the successor now understands how the work connects to other people and teams.

Also ask which tasks they can now perform themselves.

Then look at the numbers you set up at the start:

the successor’s ratings for each critical task at the close, next to the baseline, and the expert’s rating alongside;
the number of critical tasks the successor now handles alone;
for every remaining gap, a named person who holds that knowledge;
the stories, rules of thumb and updated procedures, in a shared place with an owner.

A transfer that ends with two tasks still uncovered but a named person for each is honest and useful.

One that ends with everyone feeling confident and nothing written down is the risk the program was meant to prevent.

Sources

Leonard, D., & Swap, W. (2004). Deep smarts. Harvard Business Review, 82(9), 88-97.
Nonaka, I., & Takeuchi, H. (1995). The Knowledge-Creating Company. Oxford University Press.
Kolb, D. A. (1984). Experiential Learning: Experience as the Source of Learning and Development. Prentice-Hall.
Murphy, W. M. (2012). Reverse mentoring at work: Fostering cross-generational learning and developing millennial leaders. Human Resource Management, 51(4), 549-573. doi:10.1002/hrm.21489
Lankau, M. J., & Scandura, T. A. (2002). An investigation of personal learning in mentoring relationships. Academy of Management Journal, 45(4), 779-790. doi:10.2307/3069311