Skills Transfer & Tacit Knowledge — keeping the expertise when the expert retires
D-SKT-00 | First published 2026-07-18 | Revised 2026-08-02 | 7 documents
Disclosure: Gemba Works Lab is supported by advertising. This page currently contains no affiliate links; if affiliate links are added in the future they will be clearly marked (per the FTC Endorsement Guides).
A Big Four consultancy and The Manufacturing Institute project that U.S. manufacturing could need as many as 3.8 million new workers between 2024 and 2033 — and that around 1.9 million of those roles could go unfilled. Behind the hiring statistics sits a quieter problem: when a 30-year veteran leaves, the part of their job that was never written down leaves with them. Japanese manufacturers ran into this demographic wall roughly a decade before the U.S. — the "2007 problem," when the postwar baby-boom generation began mass retirement — and developed practical, unglamorous methods for capturing what veterans know. This section translates those methods for U.S. plants of 10–500 people, written for plant managers, HR leads and owners. The organizing question for the section: when a veteran gives notice, what exactly is at risk — and in what order should you capture it?
BLS numbers on the machinist pipeline (34,200 openings a year against flat employment), plus a worked example for estimating your own exposure — with every assumption shown.
A copy-ready matrix template with a 4-level scale, the one rule that keeps it honest, and how Japanese plants use theirs to drive cross-training — not just to decorate the office wall.
What 暗黙知 actually means, the SECI knowledge-conversion cycle in shop-floor terms, and the difference between documenting a procedure and capturing judgment.
A capture method that runs on a smartphone and one day of the expert's time — including the interview questions that surface the "why," not just the "how."
A U.S. Department of Education estimate (via JLL, April 2026) says only 2 replacements enter manufacturing/trades for every 5 who retire — plus what the Manufacturing Institute's aging-workforce data means for a shop under 50 people.
A compressed 4-week sequence for shops that get 30-90 days' notice — find out what they know, document what stops the line, then shadow and reverse-shadow before the departure date.
You've built one department's matrix in 30 minutes. Here's the department-by-department sequence to roll it out shop-wide without buying software.
Where to start
Start with the skills matrix — it takes 30 minutes and tells you where your single points of failure are. Then read the retiring machinist guide to put a defensible number on the risk for your management team. The capture methods (guides 3 and 4) come after you know whose knowledge to capture first. If your improvement program keeps rediscovering that "the standard lives in Joe's head," this section pairs naturally with Kaizen in Practice.
We do not publish invented case studies or testimonials. Statistics on this page are from the cited Manufacturing Institute/a Big Four consultancy study and U.S. Bureau of Labor Statistics data; worked examples are labeled as illustrative with assumptions shown. See how this site is made.