What is knowledge transfer? Methods, examples and how to check it worked
What knowledge transfer is, which knowledge moves easily and which does not, the main methods and when each fits, and how to check that a transfer worked.
Knowledge transfer is the process by which what one person or team knows becomes usable by another. In research, the standard definition comes from Linda Argote and Paul Ingram (2000), and Argote repeats it in her 2024 review. It is the process through which one unit (a team, a department or a site) is affected by or learns from the experience of another. So you judge a transfer on the receiving side: afterwards, the receiver can do something they could not do before.
It does not happen by itself. In a study of 32 transfers of manufacturing technology inside companies (Galbraith, 1990), one in three failed and was abandoned. This guide explains which knowledge moves easily and which does not, the main methods and when each fits, the moments that need a transfer, and how to check that it worked.
In the UK, the term is also used for Knowledge Transfer Partnerships, a public scheme that links businesses with universities and research organisations. This guide does not cover them.
Which knowledge transfers easily, and which does not
In 1966 the philosopher Michael Polanyi wrote a sentence that people still quote: "we can know more than we can tell". In 1994, Ikujiro Nonaka built on this idea in a theory of how organisations create knowledge: through a constant exchange between two forms of knowledge. Explicit knowledge can be put into words and written down. Tacit knowledge comes from experience: people show it more easily than they explain it.
| Examples | How it moves | |
|---|---|---|
| Explicit | a procedure, a price list, the settings of a tool | in writing: a procedure, a work instruction, a handover note |
| Tacit | hearing that a machine is about to fail, calming an angry customer, knowing who to call at a supplier | by watching, by working next to the expert, by talking through real cases |
Writing is not enough for tacit knowledge. In her review, Argote notes that explanations in words transfer it poorly, and that it may need observation or an apprenticeship. The ISO auditing practices group says the same in its guidance on organisational knowledge (2016): for tacit knowledge, or knowledge from experience, document control is not enough to make it available and usable. The French agency for working conditions, Anact, notes the same in its guide on older workers (2009, in French): an experienced tutor does not always find it easy to explain their own skills, especially an older worker whose skills may have become second nature.
The type of knowledge is not the only barrier. Gabriel Szulanski studied 122 transfers of best practice in eight companies (1996). The main barriers were not about motivation. They were a receiver without the background to absorb the practice, a practice that nobody fully understood, and a difficult relationship between the source and the receiver.
Our article on tacit knowledge shows how to write down the part that can be written.
Knowledge transfer methods, and when each fits
The more tacit the knowledge, the more direct contact it needs. Argote's review finds that, for knowledge that is hard to interpret, methods that bring people together, or that move people between teams, work better than documents.
| Method | What it moves best | When to use it |
|---|---|---|
| Documentation | steps, rules, the reasons for a decision | rare or critical tasks, or when you need a record |
| Mentoring | judgement, contacts, how the job is really done | a new role, or skills that grow over months |
| Shadowing | how the expert works and decides, day to day | at the start of a handover, before the receiver acts alone |
| Pairing and overlap | tacit skills, relationships with customers and suppliers | a planned departure, with time when both people are in the role |
| Community of practice | what one profession learns across sites | a role spread across teams or countries |
| After-action review | the lessons of a project, an incident or a crisis | after every event that matters |
Documentation covers everything that can be written down: procedures, work instructions, handover notes. In one study that Argote reviews, in a grocery chain, using a shared document repository improved performance, especially for routine tasks. Our guide on standard operating procedures explains how to write procedures that people follow.
Mentoring, shadowing and pairing move what writing cannot, if someone organises them. Anact warns that putting an older and a younger worker together is not enough on its own. Its guide asks managers to plan time for the transfer, insists that tutors should be volunteers, and advises setting the goal and the success criteria at the start. The European Agency for Safety and Health at Work (EU-OSHA) gives a simple example in its guide on workers of all ages: hire the successor about six months before a retirement, so the knowledge can be passed on. As for results, a meta-analysis by Eby and colleagues (2008) found that people with a mentor did better on many outcomes, but the differences were usually small.
After-action reviews are the method with the best measured effect. A meta-analysis of 46 samples (Tannenbaum and Cerasoli, 2013) found that debriefs improved performance by about 25% on average, for teams and individuals alike. A later meta-analysis of 61 studies (Keiser and Arthur, 2021) also found a large effect. Our after-action review template gives a structure for the meeting, and our lessons learned template for recording what you learned.
Communities of practice are harder to evaluate. In 1991, Jean Lave and Etienne Wenger described how newcomers learn a trade by taking part in the work of a community of practitioners, first at the edges, then more fully. But in a 2023 review of 50 healthcare studies, only nine showed a statistically significant effect.
Most transfers need two methods: write down what can be written, and work side by side on the rest. The order matters too. In a study of 116 transfers in eight companies, Szulanski, Ringov and Jensen (2016) found that, if the reasons a practice works are unclear, face-to-face methods reduce difficulty most when they come early in the transfer.
When knowledge transfer is needed
In a study of 10 companies (2006, in French), Anact describes four reasons why organisations set up a transfer: to bring in new people, to adapt to a change, to support moves between roles, and to keep key skills that only a few employees hold. In practice, a transfer is needed at these moments:
- a retirement, often known years in advance;
- a resignation, or the end of a contract or a temporary assignment;
- a move to another role, or a promotion;
- work handed to an outside provider, or brought back in-house;
- a handover between teams, between shifts, or at the end of a project;
- fast growth, when new people arrive faster than experienced staff can answer their questions.
Short handovers matter too. In a 2014 study in nine hospitals, a structured handover programme for junior doctors (residents) reduced medical errors by 23%, without making spoken handovers longer.
To decide where to start, look at roles one by one. The French government department for the civil service (DGAFP) does this in its guide on knowledge transmission (2007, in French): it found that the risk depends more on the post than on the job family, and it scores both. Its criteria include the departures expected in the next three and five years, how hard the role is to recruit for, and whether the post is unique. It also looks at how many years it takes to master the knowledge, and at the practical skills that even a well-trained newcomer will struggle to learn.
People are working longer, so retirements arrive gradually. Across the European Union, 66.4% of people aged 55 to 64 were in work in 2025, up from 43.2% in 2009, according to Eurostat and the Organisation for Economic Co-operation and Development (OECD). In the UK, 72.2% of people aged 50 to 64 were in work in April to June 2026, according to the Office for National Statistics and the Department for Work and Pensions. Many of the people who will retire in the coming years are still at work, so there is time to plan the transfer.
How to check that a transfer worked
A transfer is "not an act [...] but a process", Szulanski wrote in 2000. It does not end with the last session. On measurement, Argote warns that surveys do not capture tacit knowledge. She suggests looking at the receiver's performance, or at changes in the way they work. In one factory she cites, the first shift took many months to reach a level of productivity that a second shift then reached in a couple of weeks.
Agree on a few concrete tests before you start:
- the receiver does the task alone, to the agreed standard, by the agreed date;
- the time it takes them to work alone compares well with earlier new starters;
- questions stop going to the previous role holder;
- quality holds: errors, complaints and incidents on the transferred work do not rise;
- what was written down is used and updated by its new owner.
When knowledge is not transferred, the cost shows. In a long-term study of an Australian government department (2018), the main effects of losing experienced staff included lower productivity, slower work and newcomers who took longer to become competent.
What our customers do
At Fnac Darty, seasonal hiring and outsourced teams mean new advisors must learn the job fast. Elium is used from the first day of onboarding, so trainers and advisors work from the same current procedures: more than 2,000 of them, in one place. At SPIE ICS, turnover is part of the job on the service desk. When someone leaves, what they knew stays in the team's knowledge base for the next person. Onboarding fell from three weeks to two.
Elium is built for this: what was transferred stays on one shared platform, every piece of content has an owner and a review date, and when someone leaves, their content passes to a colleague.
Common questions
It is the process by which knowledge held by one person or team becomes usable by another. Research defines it as the process through which one unit is affected by or learns from the experience of another (Argote, 2024, after Argote and Ingram, 2000). You judge it on the receiving side: what the receiver can do afterwards.
KT is short for knowledge transfer. The process has five steps: find the knowledge at risk and who holds it, name the receiver, choose methods that fit the type of knowledge, book time for the sessions, and check that the receiver can do the work alone. Our guide to a knowledge transfer plan goes through each step.
A list of what must be handed over before someone leaves a role or a service changes hands: documents, contacts, access rights, open issues, recurring tasks, and the check that the receiver can work alone. Our offboarding checklist and project handover templates are two examples.
The most common one is a knowledge transfer plan: the knowledge at risk, its holder and its receiver, the methods and dates, and the test that closes the transfer. Our knowledge transfer plan template shows one filled in for a departing senior engineer. A handover note is a shorter example: the main contacts, the role expected of the newcomer, the tools and the indicators to watch.
Tacit knowledge: practical skills, judgement, and knowing the people involved. It moves through observation, pairing and practice with the expert, more than through documents. Szulanski's research also shows that a receiver without the right background, or a strained relationship with the source, makes any transfer harder.
Where to start
Find the question in your team that only one person can answer. Write down who holds the answer, who will need it next, and how you will know it has passed to them: that is the first line of a knowledge transfer plan. To see how Elium keeps transferred knowledge current, book a demo.
Sources
- Linda Argote, Knowledge transfer within organizations: mechanisms, motivation, and consideration, Annual Review of Psychology, 2024, which discusses Argote and Ingram (2000), Galbraith (1990) and Epple, Argote and Murphy (1996).
- Michael Polanyi, The Tacit Dimension, 1966, reissued by the University of Chicago Press in 2009.
- Ikujiro Nonaka, A dynamic theory of organizational knowledge creation, Organization Science, 1994.
- Gabriel Szulanski, Exploring internal stickiness, Strategic Management Journal, 1996, and The process of knowledge transfer, Organizational Behavior and Human Decision Processes, 2000; Szulanski, Ringov and Jensen, Overcoming stickiness, Organization Science, 2016.
- ISO/TC 176 and IAF, ISO 9001 Auditing Practices Group, Guidance on organizational knowledge, edition 1, 13 January 2016.
- Anact, Acquérir et transmettre des compétences (in French), 2006, and Maintien et retour en activité professionnelle des seniors (in French), 2009.
- DGAFP, La transmission des savoirs (in French), January 2007.
- EU-OSHA, Healthy Workplaces for All Ages, Knowledge transfer, training and life-long learning.
- Eby and colleagues, Does mentoring matter?, Journal of Vocational Behavior, 2008.
- Tannenbaum and Cerasoli, Do team and individual debriefs enhance performance?, Human Factors, 2013; Keiser and Arthur, A meta-analysis of the effectiveness of the after-action review, Journal of Applied Psychology, 2021.
- Lave and Wenger, Situated Learning, Cambridge University Press, 1991; Noar and colleagues, The aims and effectiveness of communities of practice in healthcare, PLOS ONE, 2023.
- Starmer and colleagues, Changes in medical errors after implementation of a handoff program, New England Journal of Medicine, 2014.
- Massingham, Measuring the impact of knowledge loss, Journal of Knowledge Management, 2018.
- Eurostat, Employment rate of older workers, age group 55–64, updated 10 September 2026, and OECD labour force statistics.
- Office for National Statistics, Employment rate, aged 50 to 64 (LF2U), released 15 September 2026; Department for Work and Pensions, Economic labour market status of individuals aged 50 and over: September 2026.
- UK Research and Innovation, Knowledge Transfer Partnership guidance.