To prevent knowledge loss when a maintenance expert leaves, the method comes down to three moves: target what only they know, capture that knowledge during their final jobs rather than in a meeting room, and structure it into experience records the team can query once they have gone. The whole thing fits into the last six months.
The issue goes beyond the individual case: 89% of American industrial leaders acknowledge a talent shortage, and 2.4 million roles could go unfilled by 2028 [Source: Deloitte/Manufacturing Institute, 2018]. An expert who leaves is replaced slowly, and rarely at the same level. Industrial intelligence platforms such as Mimorian model the equipment and structure fault diagnosis so that maintenance teams build on their expertise with every job. This article sets out the method and the checklist for organising the handover before the final day.
Why does the classic handover fail?
The two usual handover tools quickly show their limits.
First, the generic pairing: putting a junior technician behind the expert for a few weeks captures the routine moves, the ones the team already knows. The rare fault, the one that comes round once every five years, is unlikely to occur during the pairing window.
Then the handover document: written in the final weeks, it lists generalities the technical documentation already covered. The most valuable knowledge is tacit, and the expert does not always know what they know. The sound that signals a failure, the order in which they run their checks, the component they rule out straight away: none of it strikes them as worth a line.
The cost of that knowledge evaporating can be measured: inefficient knowledge sharing costs a large company an average of 47 million dollars a year in lost productivity and repeated errors [Source: Panopto, 2018]. Our article on expertise draining away as experts retire covers this in detail.
What should be captured first?
Not everything carries the same weight. Three types of material hold most of the value:
- Rare faults. The ones that come round once every five years and that only the veteran knows how to handle. They are the ones that will cost the most once they have gone.
- The quirks of the installed base. Successive modifications, repairs that became permanent, gaps between the original drawing and the actual machine. This knowledge exists in no manufacturer documentation.
- The way of reasoning. The order in which the expert looks, what they rule out straight away, what triggers their doubt. It is the hardest to pass on and the most valuable, because it is a line of diagnostic reasoning rather than a list of facts.
Your field technicians hold this tacit knowledge day to day; an expert leaving simply makes the inventory urgent.
How can knowledge be captured within the flow of work?
Capture works better during real work than in an interview. Three practices make the difference:
Shadow the final jobs. Every call the expert takes on a machine becomes a capture opportunity: the successor observes, and above all the expert reasons out loud. The reasoning matters more than the gesture.
Structure every case as an experience record. Symptoms observed, hypotheses tested, root cause, fix applied: a structured record can be found again six months later, whereas a free-form note gets lost. That is the difference between a base you query and a folder nobody opens.
Accept voice as the interface. An expert will describe a job out loud far more readily than by filling in a form. Voice dictation, transcribed and structured automatically, captures the detail that would have been skipped in writing. We set out this approach in voice rather than data entry for the job report.
The end goal fits in one sentence: leave behind a base that the technician taking over the role can query, in their own words, at the moment the fault occurs.
The checklist for the final six months
A retirement has one advantage over a resignation: it can be seen coming. Here is the sequence, from D-180 to the last day.
- D-180: map the dependency. List the equipment for which the expert is the sole point of reference and the faults they alone handle. This risk map guides everything that follows.
- D-150: prioritise the material. Rare faults, quirks of the installed base, workarounds running in production: rank them by the likely cost of forgetting them.
- D-120: target the pairing. Arrange two-person jobs on the critical equipment identified, rather than a generic shadowing period.
- D-90: capture in real conditions. The expert reasons out loud and every job produces a structured record (symptoms, hypotheses, root cause, fix).
- D-60: have the cases replayed. The successor handles new faults using the base built so far; whatever they lack reveals the gaps to fill.
- D-30: close out with targeted interviews. Only on the cases that never came up during the period, working from the D-180 map.
- D-0: check autonomy. The final test is simple: can the successor retrace the expert's reasoning on their own, on the five most critical faults?
Frequently asked questions
How long does it take to capture an expert's knowledge before they leave? Six months covers the critical equipment and a good share of the rare cases. Below three months, the focus narrows to the faults the expert alone handles and to the quirks of the installed base.
Do you need software to capture a maintenance expert's knowledge? The medium matters less than the structure: symptoms, hypotheses, root cause, fix. A dedicated tool adds the immediate answer: the successor asks their question as the fault happens and finds the expert's reasoning, where a folder requires knowing where to look in the first place.
What if the departure is less than three months away? Narrow the scope and keep the method: map the dependency in a week, then spend all the remaining time on rare faults and on gaps between drawings and actual machines, captured in real conditions with voice dictation.
Conclusion
Three points to remember.
- The calendar is on your side. A retirement can be planned: the window of the last six months is enough when it is organised.
- Not everything carries the same weight. Rare faults, quirks of the installed base and the way of reasoning come before knowledge that is already documented.
- The flow of work beats the meeting room. Capturing during jobs produces a base you can query; the handover interview produces a document nobody reads again.
Next step: draw up the map of equipment for which a single person holds the keys. For the full picture of the approach, read our complete guide: Capitaliser le savoir-faire des experts avant la retraite.
Sources:
- Deloitte/Manufacturing Institute, 2018 : Skills gap and future of work study (89 % des dirigeants constatent une pénurie de talents ; 2,4 millions de postes potentiellement non pourvus d'ici 2028)
- Panopto/YouGov, 2018 : Workplace Knowledge and Productivity Report (47 millions de dollars par an de productivité perdue par grande entreprise)