skills transfer & tacit knowledge

How Japanese Factories Turn Tacit Knowledge (暗黙知) into Teachable Skill

D-SKT-03 | First published 2026-07-18 | Revised 2026-07-18 | Primary sources: 3

Disclosure: Gemba Works Lab is supported by advertising. This page contains no affiliate links; if any are added in the future they will be clearly marked (per the FTC Endorsement Guides).

Ask a veteran polisher how they know a surface is done and you may get: "it feels done." That answer is not evasion — it is an accurate report of how expert skill is stored. Japanese manufacturing calls this 暗黙知 (anmoku-chi), tacit knowledge: knowing held in hands, eyes and ears rather than in words. Because Japan's factories faced mass boomer retirements in the late 2000s (the "2007 problem"), their methods for converting this knowledge into teachable form are unusually well-developed — and unusually practical. This guide, for plant managers and trainers, explains the concept without the mysticism and lays out what actually transfers.

KEY POINTS

Tacit knowledge is normal, not magical — it is what any skill looks like after enough repetitions. Treating it as a veteran's mystique makes it harder to capture, not easier.
The working model is the SECI cycle (Nonaka & Takeuchi): knowledge converts between tacit and explicit through socialization, externalization, combination and internalization — each stage maps to a concrete shop-floor activity.
You cannot fully document judgment. The goal is to document enough to shorten apprenticeship, not to replace it — that honesty is what makes capture programs succeed.

Tacit vs explicit, in shop-floor terms

The distinction comes from philosopher Michael Polanyi ("we can know more than we can tell") and was developed into a management theory by Ikujiro Nonaka and Hirotaka Takeuchi in The Knowledge-Creating Company (1995), the standard academic account of how Japanese firms handle knowledge. Stripped to shop-floor terms:

	Explicit knowledge (形式知)	Tacit knowledge (暗黙知)
Lives in	Documents, programs, setup sheets	Hands, eyes, ears; pattern memory
Machining example	Feeds/speeds table for 4140 steel	Hearing chatter start two seconds before it shows on the finish
Welding example	WPS parameters	Reading the puddle and adjusting travel speed without thinking
Inspection example	The tolerance on the print	Knowing which customer's parts to check twice, and where
Transfers by	Reading, copying files	Watching, doing, being corrected — time in the presence of the skill
Walks out the door?	No, if filed	Yes — with the person

The economic stakes of that last row are not abstract: a Big Four consultancy and The Manufacturing Institute project up to 1.9 million U.S. manufacturing jobs could go unfilled by 2033, with retirements a central driver (source: The Manufacturing Institute & a Big Four consultancy, 2024). When replacement hiring is this hard, the tacit layer of your current workforce is inventory you cannot reorder — the valuation exercise is in our retiring machinist guide.

The SECI cycle at the workbench

Figure: The SECI knowledge-conversion cycle, labeled with its shop-floor equivalents.Source: adapted by the Gemba Works editorial team from Nonaka & Takeuchi, The Knowledge-Creating Company (1995).

Each quadrant is an activity you can schedule. Socialization is pairing — the apprentice absorbs by shared work (tacit to tacit). Externalization is the hard, valuable step: interviews, narrated demonstrations, video — forcing judgment into words and images (tacit to explicit). Combination merges those fragments into standards, setup sheets and training material (explicit to explicit). Internalization is deliberate practice against that material until it becomes the learner's own reflex (explicit to tacit) — at which point the cycle can run again at a higher level. Structured coaching approaches like Toyota Kata, documented for U.S. manufacturers by NIST MEP, are essentially this cycle run as a management routine (source: NIST MEP — Toyota Kata).

Four capture methods Japanese plants actually use

Paired work with narration (OJT, done deliberately). Not "shadow Joe for a while" — scheduled sessions on named jobs from the skills matrix, where the veteran must think aloud. The narration requirement is what turns socialization into externalization.
The "why" interview. A second person asks the veteran why at every decision point: why this insert, why slow down here, what were you looking at just now? Experts skip steps they no longer notice; the interviewer's naivety is the tool that catches them.
Video with commentary. Record the hands, then record the veteran watching the footage and explaining what they were sensing. The two-pass structure is the core of our 5-step video SOP method.
Skill maps with named successors. The 星取表 (star chart) from our skills matrix guide, used as the master schedule: every one-deep skill gets a capture session and a named trainee with a date.

What honestly does not transfer

A credible program admits its limits. Sensor-level skill — the calibrated hand, the ear for chatter — is built only by repetitions; no document shortcut exists, and Nonaka and Takeuchi's model never claims otherwise (internalization is a practice stage, not a reading stage). What capture does do is compress apprenticeship: the learner practices against the veteran's externalized decision rules instead of rediscovering them by scrap. Second, motivation transfers nowhere by paperwork — veterans share generously when capture is framed as legacy and respect, and clam up when it smells like "download your brain before you go." The Japanese framing is worth copying literally: the program is about honoring a career's worth of skill, in an industry that — per BLS — must replace tens of thousands of skilled production workers every year (source: BLS OOH — Machinists and Tool and Die Makers).

Set expectations accordingly: a good capture program does not make a 10-year skill teachable in a month. It makes the first three years of apprenticeship two, and it stops the worst losses cold. That is worth a great deal — and nothing more should be promised.

references

Literature reference: Nonaka, I. & Takeuchi, H., The Knowledge-Creating Company (Oxford University Press, 1995) — the SECI model.

Last updated: 2026-07-18 | Primary sources referenced: 3 (plus 1 book reference) | Spotted an error? Tell us via the contact page.