How to Protect Your Staff When You Have No Safety Budget

Key Takeaways

• The first move in protecting your staff is a decision, not a purchase, and the decisions are free.
• An incident reported to nobody is something your business already decided, quietly, by never naming who hears about it.
• Doing the free steps first is the only way to know what size of anything you would ever need to buy.

---

Search how to protect employees from violence and you’ll get a price list. Every result assumes the answer is a thing you install. That assumption is why most small businesses do nothing for another year. The only option they can see has a number on it, and the number never fits this quarter.

The moves that change what happens on your worst shift are decisions, and they cost nothing. Who is alone, and when. Who hears about it first. What the person standing there does in the ninety seconds before anyone else knows. None of that needs a vendor. All of it needs you, which is the part nobody sells.

How do I protect employees from violence this week without spending anything?

Three moves, in this order, none of them a budget line:

1. Walk one shift. Stand where your staff stands and ask what they’d do if the person in front of them turned.
2. Decide who is never alone, and when. One scheduling change, made by you, this week.
3. Name who hears about it. A person, by name, and a second person behind them.

Start with the walk, because it’s the step owners skip. Not what the handbook says. What your staff would actually do. Most owners find the answer is a version of shout and hope somebody hears, which is a plan that depends on the building being busy at the exact moment it stops being safe.

Do it at the hour you’re asking about rather than at ten in the morning when the place is full. An empty counter at closing time and the same counter at lunch are two different rooms, and only one of them is the one you’re worried about.

You have a structural advantage here that a large company doesn’t. There’s nobody to convince. No committee meets about this, no capital cycle governs it, and nothing turns your decision into a quarter. What you approve happens this week. Inside a bigger business these three moves would take three months of alignment, and half of them would die on the way.

When and where should I change the schedule first?

Start with the hours when one person is covering, and with the parts of your premises anyone can walk into. Both facts are already in your schedule and your floor plan, and neither costs a dollar to look at.

The federal survey of workplace violence, run jointly by the Bureau of Justice Statistics, the Bureau of Labor Statistics and NIOSH, found that 39% of nonfatal workplace violence happens in the afternoon. Most of it lands while you’re open and busy, which makes the after-dark version of this the smaller half of the problem. Evening runs 18% in the same survey and overnight 9%, and those are the hours you’re most likely to have one person covering.

The same survey looked at where. Only 12% of the violence against retail sales workers happened in a restricted area, meaning somewhere access is limited or blocked, against 49% for workers overall. Roughly seven in eight incidents against those workers happened in space the public could walk straight into. The exposed part of your operation is the part you designed to be open.

So the cheapest change available to you is overlap:

• Thirty minutes at the end of the day when two people are still on.
• A second person present at open and at close.
• Cash counted at a time when the room holds more than one person.

None of that is a purchase, and all of it moves the specific hour you’d otherwise be gambling on.

Who hears about it when something happens?

Somebody at your business has to be the person an incident gets told to, and they need a name. If you’ve never named anyone, it’s you by default, and your staff has quietly worked out whether that’s a real option.

This is the highest-value free move you have. The same federal survey put the share of nonfatal workplace violence reported to police at 39%, which leaves about 61% that never reached them. Whatever your instinct says about how much of this you’d hear about, most of it goes nowhere official.

The reasons matter more than the number. Among people who skipped the police, the largest group, 39%, had reported it to another official instead. The survey never asks whether anyone told an employer, and the officials it counts are guards, building managers and school staff. That’s an institutional layer, and a twenty-person business has none unless the owner builds one.

Underreporting is a rational response rather than a training gap. Ask most workers who they’re supposed to turn to and the honest answer is nobody in particular. The ones who do raise it often find it becomes their word against someone else’s. People read that situation correctly and stay quiet, and a poster about speaking up doesn’t change the math.

What changes it is naming two people rather than one:

• The first is whoever your staff would actually go to, which may not be you.
• The second exists for the case where the first person is the problem. That’s the case a single named contact can never cover.

Then say what that person does when they’re told, because a name with nothing behind it collects the report and stops there. Three lines is enough. Who they call, whether the shift keeps running, and what gets written down while it’s fresh. The last one matters more than it sounds. A threat described the same day is a different document from one reconstructed for an insurer four months later. You’ll only ever have the version somebody wrote at the time.

What actually requires money?

One thing: getting help to a specific room while somebody is still standing in it. Everything above this line is a decision. That’s where a decision stops being enough.

A system the owner is sold on and the staff don’t trust has already failed, and it fails quietly. Trust has to hold at every layer, from the equipment to whoever responds to you as the employer. Break it anywhere and people stop using the thing, and nothing tells you it’s happening.

That’s why the free steps come first, and money is only part of the reason. Decisions look different made from the position of the person who needs protecting rather than from the desk. Once you’ve stood on that shift and named your two people, you know three things you didn’t know before:

• How many of your staff are ever the only one there.
• Which rooms have no line of sight from anywhere else.
• Which hours are genuinely thin.

Those three facts are the specification. Walk into a purchase without them and you’re buying by catalog. Which route makes sense for a business your size, and what the alternatives are worth, is a separate question that deserves its own answer.

Whatever you eventually put in, the people using it have to have practiced it. Equipment that arrives in a box and gets mentioned once at a staff meeting is equipment nobody reaches for under pressure.

Nothing here promises fewer incidents. Nobody has measured whether any of it lowers the rate at a business your size, and anyone telling you otherwise is selling. What these moves change is who knows, how fast, and what happens next.

The price list you started with was the wrong first page. It’s the last one. Before it sit three decisions that cost you an afternoon, and they’re yours alone to make, which is the same fact that makes them possible this week.

So start with the shift you already worry about. You know which one it is. To protect employees from violence on that shift, name who is never alone, name who hears about it first, and tell your team both before Friday. When you do eventually talk to somebody selling a system, you’ll have the question worth asking. How does the person in the room call for help without letting the other person in that room know they did?

References

1. Bureau of Justice Statistics, Bureau of Labor Statistics and NIOSH, Indicators of Workplace Violence, 2019 (NCJ 250748; NIOSH 2022-124), tables 6.2, 6.3, 7.1 and 7.5.