How to Reduce Security Guard Overtime Without Leaving Posts Uncovered

It is 11pm, payroll is due tomorrow, and the number on the screen is bigger than the schedule you built. You know roughly how it happened. You just cannot point to the exact shift.

Here is how to reduce security guard overtime without leaving a post uncovered: know what federal law actually requires, understand the patterns that create overtime in guard companies specifically, and build a schedule and a weekly check that flag a guard approaching 40 hours while there is still time to swap the shift, not after payroll already ran.

What counts as overtime for a security guard, under federal law

The baseline comes from the Fair Labor Standards Act, the federal law that governs overtime pay. For a guard who is a non-exempt hourly employee, which is how most people working a post are classified, the rule is simple: hours over 40 in a single workweek get paid at one and a half times the regular rate.

The part that trips people up is the workweek itself. The regulation is titled Each workweek stands alone, and it says the law does not permit averaging of hours over two or more weeks. Its own example is the one that matters to you: 30 hours one week, 50 the next. That is overtime in the second week, even though the two weeks average out to 40. If you run payroll biweekly, your pay cycle and the legal unit of measurement are not the same thing.

The other trap is coverage across sites. The law puts the duty on the employer for the employee's hours in the workweek. It does not attach to a job site. If the same guard works two of your client accounts in the same week, those hours combine toward the 40 hour threshold. One guard, one employer, one workweek, one total, not a separate allowance per site.

Owners often assume that paying different rates at different sites separates them. It does not. Federal regulation covers exactly that case and requires the regular rate for the week to be the weighted average of those rates, calculated across the combined hours.

One more thing about "the same employer". This aggregation is about one company's own sites. If a guard also works for a different company, those are separate employers and separate calculations, which is a different question with its own rules.

Everything above is federal law. It is the floor, and it applies to every guard company in the country. It is not always the whole answer. Some states add rules on top, and a few of them change the shape of the problem rather than just the numbers: a handful require overtime after a set number of hours in a single day, not just after 40 in a week, which means a guard can earn overtime in a state like that without ever reaching 40 hours. If you operate anywhere with its own wage and hour rules, work out your state's position alongside the federal baseline, do not assume one covers the other.

This is general information and not legal advice. Confirm anything with real money attached against the U.S. Department of Labor, or an employment attorney. If Florida licensing is also on your plate, see our guide to the Class D license.

Why overtime sneaks up on small guard companies

Nobody plans overtime. It arrives in small, reasonable decisions that add up by Friday.

A guard calls off two hours before a shift starts. You need the post covered now, so you call the guard who already worked a full day, because he says yes. That is overtime, created in about ninety seconds of relief that the post is covered.

Guards also arrange coverage among themselves, which is usually fine, until two guards trade shifts and nobody updates the master schedule. The company finds out what actually happened when the timecards come in.

Then there is the multi-site problem. If guards rotate across two or three client accounts, whoever schedules Site B often has no idea how many hours that guard already picked up at Site A. Two individually reasonable schedules combine into overtime nobody assigned on purpose.

Biweekly payroll hides the rest. A heavy week and a light week average into a total that looks fine, right up until you remember the law measures the week you worked, not the total.

None of this is a discipline problem. It is a visibility problem. The information exists, but it is scattered across whoever built each shift, and nobody sees the running total until it is too late to change it.

The real cost of letting it slide

Guard work is usually billed to the client at a flat hourly rate, agreed months or years ago and rarely renegotiated. Overtime does not renegotiate itself alongside it.

Every overtime hour your company pays but cannot bill to the client comes straight out of your margin on that account, not out of some abstract overhead line. Same work, same billed rate, costing you more to deliver than it did the week before.

That matters more in this business than most, because a guard company is mostly labour. There is not much else in the cost base to absorb hours nobody planned. An account that looked profitable when you priced it can quietly stop being profitable, one uncovered shift at a time, without anyone deciding that it should.

The overtime itself is rarely the failure. Someone covered a post that needed covering, which is the job. The failure is not seeing it coming in time to make a different choice, like calling the next guard on a bench list instead of the one already on the clock.

How to build a schedule that does not create accidental overtime

The fix starts with a change in what you build the schedule against. Most weekly schedules get built shift by shift, filling one gap at a time until the week looks covered. That makes every individual assignment look reasonable while missing the number that actually matters: each guard's running total for the week.

• Build against weekly hours, not open shifts. Before assigning anyone's next shift, know how many hours they already have on the books for that workweek, across every site they cover for you. A guard is not whoever is free Thursday night. A guard is a running total that happens to have Thursday night open.
• Treat one guard as one number. If Site A books him for 25 hours and Site B books him for 20 more without checking, that is 45 hours and a surprise, not two clean weeks. Whoever builds the schedule needs the combined total, not their own site's slice of it.
• Flag the 40 hour line before you assign, not after. This is the step that gets skipped, because the flag usually happens after the shift is worked, when someone reviews timecards and notices. By then the overtime already happened and all that is left is paying for it.
• Keep a bench list for call-offs. When a shift opens at the last minute, the easy move is calling whoever is already on shift and asking them to stay late, and that is exactly how one call-off turns into overtime. Two or three guards who are usually available and kept under 40 hours give you a first call that does not automatically cost one and a half times.

Catching it before the assignment means you get to choose: swap the shift, split it, or knowingly approve the overtime because the coverage is worth it. Finding out on payday is not a choice. It is a bill.

Catching overtime before it happens, not after payroll

Everything above needs one accurate, current number: how many hours each guard has actually worked this week, not how many the schedule planned. Those two numbers drift apart constantly, through late clock-ins, shift trades, and the occasional generous timesheet. A schedule tells you the plan. It does not tell you what happened.

This is the gap Apeiron CAD is built to close. Guard shifts are tracked with GPS-verified time records, so the hours on file reflect when a guard actually clocked in and out, not when they were scheduled to. Location is collected only while a guard is clocked in. It is visible to the guard whenever it is active, it stops the moment they clock out, and the coordinates are deleted after 30 days. It is a record of time worked, not a leash.

The live map shows every clocked-in guard at once, across every site, with clear on-site and off-site status. Whoever builds next week's schedule can see a guard's real running total, combined across every site that guard covers, before deciding who gets the next open shift. The 40 hour line stops being something you discover after the fact and becomes something you can see coming with days to spare.

When payroll runs, the payroll-ready CSV export reads those same clock records rather than the original plan, shaped for the imports QuickBooks and Gusto accept. It is the same shift data that already builds your daily activity reports. What gets paid is what was worked. That will not eliminate overtime, and some of it is worth paying for. It removes the version that happens because nobody could see the total until it was already on a timecard.

A simple weekly overtime check

You do not need software to do this, but you do need to do it on a fixed day, midweek, before next week's schedule locks.

1. Pull each guard's hours worked so far this workweek, combined across every site they cover for you.
2. Flag anyone already past 30 hours who still has shifts scheduled this week.
3. For anyone on more than one site, check the combined total, not each site's number separately.
4. Check that next week's draft is not just a copy of this week's, repeating whatever overtime already happened.
5. For anyone flagged, make the call before the shift is assigned: swap it, split it, or approve the overtime on purpose.

Five steps, on the same day every week rather than whenever someone remembers. The value is not in any single step. It is in doing it before next week's schedule gets built on top of this week's numbers.

Common questions about security guard overtime

Is overtime calculated weekly or biweekly?

Weekly. Federal law measures overtime against a single, fixed workweek, not your pay cycle, and it does not permit averaging hours across two or more weeks. A heavy week and a light week never cancel each other out.

Do guards get overtime for working two different sites in the same week for the same company?

Yes. The obligation attaches to the employer and the workweek, not to the job site, so hours worked across every one of your sites count toward the same 40 hour threshold. You cannot treat each account as its own job with its own 40 hour allowance.

Does paying different rates at different sites change the calculation?

It changes the rate, not the threshold. Where an employee works at two or more different rates for the same employer in one week, federal regulation requires the regular rate to be the weighted average of those rates, applied across the combined hours.

What about my state?

Federal law is the floor and it applies everywhere, but it is not always the whole answer. Some states add their own rules on top, including a few that require overtime after a set number of hours in a single day rather than only after 40 in a week. In a state like that, a guard can earn overtime without ever reaching 40 hours, which the weekly math above will not catch. Check your state's rules alongside the federal baseline rather than assuming one covers the other.

None of this makes overtime disappear, and it should not. Some of it is the fair cost of covering a post that needed covering. What changes is when you find out. Building the schedule against real running totals, checking it midweek, and seeing hours accumulate turns overtime from a number you discover on payday into a decision you get to make in advance.

Check the source

Requirements in this post were checked against federal law and regulation in August 2026 and are provided as general information, not legal advice. Wage and hour rules change, and states add their own. Before acting on any of it, confirm directly with the U.S. Department of Labor. If a specific overtime question carries real exposure for your company, ask an employment attorney who practices in this area rather than a software company's blog.

Related reading: the Florida Class D license owner's guide, and how to write a daily activity report.