How Many Employees Does One 24/7 Post Need in the United States?
Holding one position every hour of 2026 in the United States takes 8,766 staffed hours. One full-time employee is genuinely available for about 1,961 of them, so the post needs 4.47 employees, which is 5 whole people.
One ring per person, one year each
Each ring is one employee's year, broken wherever they are on holiday, on leave, or off sick. No single ring closes, which is the whole problem: one person cannot hold a post that never stops.
Stack 5 of them and the picture closes. Trace any line out from the center and it crosses at least one solid arc, so every hour of 2026 has somebody on it. That is the relief factor, drawn rather than asserted.
Where the number comes from
- 1
Staffing one position 168 hours a week comes to 8,766 hours across the year.
- 2
The United States has 261 weekdays in the year, the starting point for one employee.
- 3
Take out 10 public holidays that fall on a weekday, 0 days of statutory paid leave, 2.9 days of average sickness absence, and 3 days of training: 15.9 days lost.
- 4
That leaves 245.1 working days per employee, and at a 40-hour week that is 8 hours a day, so 1,961 available hours a year.
- 5
8,766 coverage hours divided by 1,961 available hours is a relief factor of 4.47, so the post needs 5 people.
| Weekdays in the year | 261 |
| Public holidays on a weekday | −10 |
| Statutory paid annual leave | −0 |
| Average sickness absence | −2.9 |
| Training and onboarding (assumed) | −3 |
| Working days available | 245.1 |
| At 8 hours a day | 1,961 hours |
the United States gives no legal right to paid time off on public holidays. Time off is near-universal in practice, so the 10 weekday holidays are still deducted, but that is a reading of custom rather than of law.
The training allowance of 3 days is a planning assumption rather than a sourced statistic, and it is deliberately low. Every other figure on this page carries a source. If your operation runs heavy certification cycles, raise it and the relief factor rises with it.
What the usual rule of thumb gets wrong here
The common shortcut is to divide coverage hours by contracted hours and add a quarter for absence. In the United States that gives 5.25 against a derived answer of 4.47, so the shortcut is high by 17.4 percent. A flat buffer travels badly because the things it stands in for, statutory leave and the length of the contracted week, are set nationally.
Nobody ever away. Not a real answer.
The usual shortcut, over by 17.4%.
From the United States's own leave, hours, and absence.
Other coverage shapes
The same availability figure answers any coverage window. Only the hours the post must be staffed change.
| Coverage | Hours a week | Relief factor | People |
|---|---|---|---|
| Round the clock, every day | 168 | 4.47 | 5 |
| Round the clock, weekdays only | 120 | 3.19 | 4 |
| Two shifts, every day | 112 | 2.98 | 3 |
| Extended day, weekdays only | 60 | 1.6 | 2 |
If your contract is more generous
0 days is the statutory floor in the United States, and most employers sit above it. Every extra week of leave is another week the post has to be covered by somebody else.
| Annual leave | Relief factor | People per post |
|---|---|---|
| 0 daysstatutory floor | 4.47 | 5 |
| 20 days | 4.87 | 5 |
| 25 days | 4.98 | 5 |
| 28 days | 5.05 | 6 |
| 30 days | 5.09 | 6 |
| 33 days | 5.17 | 6 |
The inputs, and what they do not say
How the leave figure is defined
There is no federal statutory paid annual leave, so the figure is 0 working days on a 5-day week. It excludes public holidays, and in the US paid public holidays are also not a legal entitlement for private-sector workers, so nothing is being netted out here.
Why the contracted week is not the legal maximum
This page divides by a 40-hour week, the typical full-time contract in the United States, not by the 40-hour statutory figure. The two match here, so the distinction makes no difference to the answer.
What these figures cannot be used for
Not a ranking. The sickness figure here comes from a national statistical source, while other countries in this library use the OECD labour force survey. Those systems define absence differently, and self-reported survey figures undercount short spells that administrative records capture. Each number is the best available answer for planning inside its own country, and the gap between any two of them is partly a gap between statistical methods rather than between workforces.
A relief factor also says nothing about whether a roster is legal or humane. It sets how many people the post needs across a year; it does not check rest periods between shifts, consecutive nights, or weekend distribution. Those constraints decide whether the pattern you build on top of this headcount is one anybody can actually work.
Research notes on United States
Weekly hours: 40 is a pay threshold, not a legal ceiling. The FLSA requires overtime at 1.5x after 40 hours in a workweek, and DOL states there is no limit on the hours employees aged 16 or over may work in a workweek. Some states add daily overtime (for example California after 8 hours a day), and a few sectors have their own rules, so treat 40 as the point where cost changes rather than where scheduling must stop.
Collective hours: the US has no sector-wide collective agreements that shorten the week. Only 11.2 percent of wage and salary workers were represented by a union in 2025 (10.0 percent were members), so there is no national bargained norm below 40. Two caveats for a planner. First, the conventional contracted full-time week is 40 hours, which is why collectiveWeeklyHours repeats the statutory number rather than falling below it. Second, actual hours run above it, not below: CPS 2025 annual averages show people who usually work full time in nonagricultural industries averaged 41.8 hours actually at work per week. If the calculator models contracted hours use 40, if it models hours actually worked use 41.8.
Annual leave: zero is correct and is the figure most often got wrong. The FLSA does not require payment for time not worked, including vacation, sick leave, and holidays, and DOL states these are matters of agreement between employer and employee. Three specific traps. (1) The 11 federal holidays apply to federal employees and federal offices, they are not a private-sector entitlement, so do not add them and do not report a leave figure "including public holidays". (2) BLS employer-provided vacation data is practice, not law: in March 2025, after one year of service, 31 percent of private industry workers got 10 to 14 vacation days and 31 percent got 5 to 9. Those numbers belong in a "typical employer offers" field, never in a statutory minimum field. (3) State law does not fill the gap for annual leave: no state mandates paid vacation. Maine and Nevada mandate general-purpose paid leave usable for any reason, both capped around 40 hours (5 days) per year, and roughly 18 states plus DC mandate paid sick leave, which is a separate category from annual leave. If the calculator is used state by state, 0 is the correct national floor but Maine and Nevada users should see 5 days.
Sickness: derived, not published as days. BLS CPS Table 46 (2025 annual averages) gives a lost worktime rate for own illness or injury of 1.1 percent, defined as hours absent as a percent of hours usually worked, for full-time wage and salary workers. Conversion on a 5-day week: 0.011 x 260 scheduled working days = 2.86, rounded to 2.9 days. The same 1.1 percent appears in the 2024 annual averages, so the figure is stable. Caveats: the rate is published to one decimal, so the true value carries roughly plus or minus 0.1 day; the CPS counts an absence only when a full-time worker worked fewer than 35 hours in the reference week, so part-day absences inside an otherwise full week are invisible and 2.9 should be read as a floor; self-employed workers are excluded; and BLS warns that 2025 annual estimates are 11-month averages excluding October (data were not collected during the federal shutdown), so 2025 is not strictly comparable with other years. Do not swap in the 7 to 8 sick days figure common in HR-vendor content, that is self-reported use of employer sick-leave allowances and is measuring something else.
Sources
Every figure above except the training allowance was checked against one of these.
40-hour workweek as the FLSA overtime threshold for covered nonexempt employees
No federal limit on hours worked per week for employees aged 16 and over
U.S. Department of Labor, Wage and Hour Division, Fact Sheet #222026
No statutory paid annual leave: FLSA does not require payment for time not worked, such as vacations, sick leave, or holidays
Lost worktime rate for own illness or injury, 1.1 percent of hours usually worked, full-time wage and salary workers (basis for 2.9 days per year)
U.S. Bureau of Labor Statistics, Current Population Survey, Table 462025
Same lost worktime rate of 1.1 percent for illness or injury in the prior year, confirming stability
U.S. Bureau of Labor Statistics, Current Population Survey, Table 46 (2024 annual averages)2024
Average 41.8 hours actually worked per week by people who usually work full time in nonagricultural industries
U.S. Bureau of Labor Statistics, Current Population Survey, Table 222025
Union representation 11.2 percent and union membership 10.0 percent of wage and salary workers, showing no sector-wide bargained shorter week
U.S. Bureau of Labor Statistics, Union Members news release2025
Employer-provided paid vacation distribution (31 percent of private industry workers get 10 to 14 days after 1 year), practice rather than statute
U.S. Bureau of Labor Statistics, National Compensation Survey / Employee Benefits Survey2025
Maine earned paid leave: 1 hour per 40 hours worked, up to 40 hours per year, usable for any reason
Nevada paid leave under NRS 608.0197: 0.01923 hours per hour worked, use may be limited to 40 hours per benefit year
Common questions
- How many people does one 24/7 position need in the United States?
- 5 people. One position staffed every hour of the year is 8,766 hours, and one full-time employee in the United States is actually available for about 1,961 of them once 15.9 days of holidays, leave, sickness and training come out. That is a relief factor of 4.47, which rounds up to 5 whole people.
- What is a relief factor?
- The number of employees needed to keep one position filled continuously, once the days nobody works are taken out. It is coverage hours divided by the hours one employee is genuinely available. In the United States that is 8,766 divided by 1,961, or 4.47. A factor below the headcount you actually employ means the post runs on overtime.
- Why is a flat 25 percent absence buffer not enough for the United States?
- Because it overshoots here. A raw hours calculation gives 4.2 employees, and adding a flat quarter takes that to 5.25, against a derived answer of 4.47. That is 17.4 percent too high, because the rule of thumb assumes more leave and sickness than the United States actually loses.
- Does the statutory leave minimum reflect what employers actually give in the United States?
- It is a floor, not a norm. This page computes on the statutory minimum of 0 working days because that is the figure the law fixes and a source can be attached to. Most employers sit above it, and the sensitivity table on this page shows what the relief factor becomes at more generous entitlements, so a planner can find their own contract rather than taking the floor as the answer.
Your next schedule could take 2 minutes.
Import your team, set your rules, hit auto-fill. Most teams are live the same day.
30 days free ยท No credit card required
Already have an account?Sign in
