Before You Hire, Measure Who's Overloaded

1.9 million manufacturing jobs may go unfilled by 2033. The capacity you're recruiting for is often already on payroll. Here's how to find it.

Andrew S. Collier  ·  Founder & Managing Principal, Collier & Associates  ·  September 30, 2026

Deloitte and the Manufacturing Institute project 3.8 million manufacturing jobs to fill between 2024 and 2033, with 1.9 million at risk of going unfilled. Their September 2026 study looks to AI to widen the pool, identifying roughly two million technicians in adjacent industries whose skills could transfer. Meanwhile NAM's Q2 survey has full-time employment growing 1.0% over the next year. Hiring is the slowest lever in the plant.

There's a faster one, and most operations haven't pulled it: find out who is actually overloaded and who isn't.

Deloitte and The Manufacturing Institute projection, 2024–2033; NAM expected employment growth, Q2 2026

The over-consumption signal

A time-driven model compares each role's documented work against its practical capacity. When consumption exceeds practical hours, the role is running over. That's not a productivity compliment; it's a retention and error-rate risk, and it's invisible in headcount reports because the person is simply working through lunch and staying late.

At Volvo Penta's Americas division, two weeks of fieldwork in December 2007 surfaced one order-entry role consuming twice its theoretical capacity — an unsustainable load that no org chart showed. The same model found kit and matrix master data in five repositories and core JD Edwards functions unused while Excel carried the process. More than $700,000 in immediate savings identified, validated against SG&A within 5%.

The under-consumption signal

The opposite reading matters just as much. When a role sits under practical capacity and its time is concentrated in non-value-added work, you have released capacity: hours that can absorb growth or be consolidated.

At a $450M FDA-regulated infant-formula plant preparing for sale, the model banded consumed-to-practical hours by role and shift under explicit criteria — 90–120% consumption, all time in non-value-added or necessary-but-non-value-added work, activity covered by more than one person on the same shift. Eight FTEs qualified for reallocation, by name and shift. The same model found essential Parts Store and QA roles uncovered on nights and weekends, causing line downtime: capacity misallocated in both directions at once.

Over-consumption priced — Volvo Penta (2007) and PBM Nutritionals (2009)

Four readings, four actions

Cross the value assessment with the capacity variance and every role falls into one of four cells:

	Under practical capacity	At or over practical capacity
High non-value-added	Consolidation case — the cleanest cost-out	Overtime spent on waste — fix it for retention, not savings
Low non-value-added	Growth headroom — quantified, no action needed	Structural under-resourcing — add a resource or automate

The hiring decision belongs in the bottom-right cell only. In my experience, most requisitions are written for roles in the other three.

Value assessment × capacity variance — the cross-tab that decides whether you hire, consolidate, or leave it alone

Before the requisition

Measure first. Released capacity converts to cash through six paths — overtime and temps first, then the hire you don't make, redeployment to value-added work, retired systems and licenses, consolidation on attrition, and only then restructuring. Four of the six produce hard dollars with no involuntary separation, and the model shows which path fits which role. A single-process diagnostic takes two weeks and tells you, role by role, whether you're short of people or short of a process.

If you run a plant, a distribution operation, or a portfolio and want one process measured in two weeks, call 502-317-8900 or write to andy@andrew-collier.com. I'll show you the method using anonymized data and real examples from past engagements before you spend anything.

Andrew S. Collier  ·  Founder & Managing Principal, Collier & Associates  ·  Greater Louisville, KY

502-317-8900  ·  andy@andrew-collier.com  ·  andrew-collier.com  ·  linkedin.com/in/andrewscollier

Creator of BAS™, the task-level cost, capacity, and value model behind the BAS™ Value Creation Sprint. 40 years in enterprise transformation; 100+ enterprise systems engagements; 40+ business transformations; 29 JD Edwards implementations; six interim CIO, CTO, and IT Director roles; 6+ years of PE M&A transactions as advisor to PwC, Deloitte, and Crosslake Technologies.

Sources: Deloitte and The Manufacturing Institute, "Taking Charge" (2024) and AI technician study (September 10, 2026); NAM Manufacturers' Outlook Survey, Q2 2026; Collier & Associates engagement record (Volvo Penta, 2007; PBM Nutritionals, 2009).

Permalink: andrew-collier.com/insights/before-you-hire