Think of the person on your floor who has been there longest. When a new hire wants to know how things actually work, that is who they ask. Your succession plan lists that person's job. It does not list what they explain.
Most retirement planning in manufacturing is built around replacing a role. The harder loss is the explaining the role was quietly doing for everyone else.
Why do new hires leave after a veteran retires?
Because the veteran was doing a job no one assigned. In a plant that runs on unwritten rules, the longest-serving person is where new hires learn what the rules are. When that person retires, the rules stay and the explanation leaves. New hires who can't decode them alone do extra work to keep up, then go.
Michael Polanyi named the problem in 1966: "We can know more than we can tell." Tacit knowledge doesn't transfer through a binder. It transfers through someone patient enough to answer the same question for the fifth time.
So two problems that get reported separately, experienced people retiring and new hires not staying, are often one problem with a delay between them. The first shows up in a succession plan. The second shows up months later as ordinary early turnover, and no report connects them.
How big is the problem?
Bigger than most retention reports show, because it lands as routine first-year turnover. The Manufacturing Institute and Deloitte estimated in April 2024 that manufacturers could need as many as 3.8 million additional employees between 2024 and 2033, with up to 1.9 million positions going unfilled. Every hard-to-fill seat raises the cost of losing a new hire.
Those figures establish scale. They are too large to act on. The number you can act on is already in your HRIS.
Here is the shape of it, with the assumptions labeled. Say a 400-person plant hires 60 people a year, and a third of them land on teams where a long-tenured person recently left. If first-year turnover runs 20 percent on most teams and 35 percent on those, that is three extra departures a year. Those rates are illustrations, not benchmarks.
Three is a small number. It is also three hires you made twice, and each replacement walks into the same unexplained rules as the person before.
What actually fixes it?
Write down what the veteran was explaining, before the veteran leaves. Procedures are usually documented already. The judgment rarely is: what "done" means, which priority wins when two collide, who decides what, and what a supervisor actually expects. Explicit systems keep working after the person who carried them retires.
Most knowledge-capture efforts aim at the layer that was never the problem. They document procedures and skip the unwritten rules.
The same Manufacturing Institute report found employees were 2.7 times less likely to leave within a year when they felt they could acquire the skills they needed. A new hire can only acquire what someone has made visible.
I came to this from the other direction. I couldn't read implicit systems, so in the organizations I ran I wrote them down: promotion criteria published before anyone was measured against them, agendas sent ahead, decisions recorded with the reasoning. I did it because I needed it. It turned out to work better for almost everyone.
How do you check your own plant?
Compare first-year turnover for new hires who joined a team soon after a long-tenured employee left against everyone else. If the first group leaves faster, the person who retired was carrying unwritten rules the team still depends on. The data is in your HRIS, and the cut takes an afternoon.
- Pull 24 months of separations with hire date, team and supervisor.
- Mark every departure of an employee with fifteen or more years of tenure, by team and date.
- Split new hires into two groups: those who joined a team within twelve months of a marked departure, and everyone else.
- Compare first-year departure rates between the two groups.
- For the teams with the widest gap, ask the remaining veterans one question: what do new people get wrong in their first month? Write down the answers. That list is the first draft of your explicit system.
Where this stops: on small teams, some of the gap will be chance, and some of it is the labor market. And not every veteran was a translator; a few were the reason people left. Below a few dozen hires a year, read the result as a reason to ask questions, not a measurement.
You now have a way to see whether retirements are turning into turnover months before the turnover report would. Very few plants have run that cut. See where training does not reach.
Who on your floor gets asked the most questions by new hires?
If experienced people are retiring and new hires aren't staying, read about the handoff, or price your turnover gap with your own numbers.
Sources
- Michael Polanyi, The Tacit Dimension (University of Chicago Press, 1966).
- The Manufacturing Institute and Deloitte, "Manufacturers Need as Many as 3.8 Million New Employees by 2033", April 3, 2024.