Employer guideRead at source, 9 September 2026

Understaffing: How to Tell Whether You Actually Are, What It Does to the People Who Stay, and What Breaks the Loop

Understaffing is when the work you have scheduled is larger than the people available to do it — measured, not felt. Three numbers settle it: the share of scheduled hours you could not cover, overtime as a share of paid hours, and your quit rate against your own sector’s. The surprise in the 2026 federal data is that the labour market is not the explanation. The job openings rate is running at 4.4 percent against 4.5 percent in 2019, quits at 2.0 percent against 2.3 percent, and average weekly overtime in manufacturing at 3.97 hours against 4.23. By every published measure, hiring conditions are easier than they were before the pandemic. So if your schedule is short, the cause is almost certainly inside your own building: a replacement rate you are not keeping up with, a fill time longer than your turnover cycle, or a schedule nobody wants. All three are fixable, and the order matters.

What understaffing is, and what it is not

Being short today is not understaffing. Someone calls in sick, a truck is late, a shift runs over, and a manager covers. That is variance and every operation has it. Understaffing is structural: the schedule cannot be covered by the people on the roster even when nothing goes wrong. The distinction matters because the fixes are opposite. Variance is absorbed by float, cross-training and a call list. Structural shortfall is only fixed by adding people or by removing work, and no amount of scheduling cleverness substitutes.

The other thing understaffing is not is a headcount number. Two operations with identical rosters can be in completely different positions depending on turnover: the one replacing a quarter of its people every year is running to stand still, and the one replacing a twentieth is not. That is why the test below asks for a quit rate and not just a headcount.

Three numbers that settle it

Run these on one ordinary week — not your worst week, which is what everyone reaches for. A coverage gap above about two percent of scheduled hours is real rather than noise. Overtime above roughly eight percent of paid hours means the schedule is being carried by the people you already have. And a quit rate meaningfully above your sector’s says the problem is the back door rather than the front.

The market is not your explanation

Almost every article on this subject opens with a labour shortage. In 2026 that framing does not survive contact with the data. Openings and quits both spiked in 2021 and 2022, and both have since come back to — or below — where they sat in 2019.

Sources: BLS Job Openings and Labor Turnover Survey, total nonfarm, seasonally adjusted; BLS Current Employment Statistics average weekly overtime hours of production and nonsupervisory employees in manufacturing. Both read 9 September 2026. 2026 figures are averages of January to July for JOLTS and January to August for overtime.

This is not an argument that hiring is easy. It is an argument that the conditions are roughly what they were in 2019, so if 2026 feels much harder than 2019 did, something changed at your end and not at the market’s. That is better news than a shortage, because the things at your end are things you control.

The loop, and why it does not resolve on its own

Understaffing is rarely a level. It is usually a loop, and it runs in one direction. A seat opens. The schedule still has to be covered, so the hours go to the people who are already there, as overtime or as a sixth day. Those people are, by definition, your most experienced and least replaceable. Some of them leave. Now there are two seats and the same schedule.

The three causes, in the order they are usually true

In the frontline operations we work with, structural shortfall almost always traces to one of three things, and it is worth diagnosing in this order because the fixes get progressively more expensive.

1. You are filling slower than you are losing. If it takes you six weeks to replace someone and you lose someone every four, the gap widens no matter how good your ads are. This is arithmetic, and it is the most common answer. The fix is fill speed, not budget.
2. The schedule itself is the vacancy. Some seats are open because of what they are, not because nobody applied: the shift that starts at eleven, the weekend rotation, the site with no parking. These do not respond to more advertising of the same ad. They respond to changing the shift or paying for it explicitly, which we cover on how to fill open shifts.
3. Your replacement rate is the real number. If your quit rate is running well above your sector’s, everything upstream is refilling a bucket with a hole in it. Hiring faster still helps — it stops the gap widening while you fix the hole — but on its own it is a treadmill.

What to do, in order

1. Measure the week, not the crisis. Run the three numbers above on an ordinary week and write them down. Repeat monthly. Two data points beat any benchmark we could give you.
2. Put a number on the gap. Uncovered hours and overtime premium are both real money, and finance will fund a fix for a number in a way it will not fund a fix for a complaint. Our cost of vacancy calculator does that arithmetic.
3. Shorten fill time before you raise budget. Time to fill is the multiplier on every other number here. Most of the delay in frontline hiring is between application and first human contact, and it is free to fix.
4. Advertise continuously for the roles you always lose. If a role turns over predictably, campaign for it continuously rather than reacting to each resignation. The reactive version pays a premium for urgency every single time.
5. Fix the schedule for the seat nobody takes. If one specific shift is chronically open, it is a product problem, not a marketing problem.
6. Then look at retention. It is last on this list only because it is slowest, not because it matters least. If the quit rate is the binding number, nothing above it holds.

Where the advertising fits, honestly

Advertising moves exactly one of the three causes: how fast a seat gets refilled. It does not change your schedule and it does not change why people leave. What it does do is stop the loop widening while you work on the other two, and that is worth real money when the gap is being covered with overtime at time-and-a-half. Across the 891 campaigns in our 2026 Social Job Advertising Benchmark, the median campaign produced an applicant for $13.88, with the volume-weighted average at $8.02 — set that against a single week of premium hours and the comparison usually makes itself.

Frequently asked questions