Stop Work Authority Is Easy to Write Down and Hard to Use

Chris Fredette
Jul 30
4 min read

Every orientation I have ever sat through includes the line: anyone on this site has the authority to stop work. Everybody nods! Then a year goes by on a jobsite with two hundred people and it gets used twice.

The gap between the policy and the practice is not that people do not care. It is that stop work authority costs something immediate and visible, while the benefit is an incident that does not happen, which nobody sees and nobody thanks you for.

Why the call is genuinely hard

When you stop a task, several things happen at once. The crew stands around. The foreman has to explain the delay. Somebody's schedule slips, and on a job with liquidated damages that number is real money. If you are the sub, you have created a problem for the general contractor. If you are the safety person, you have just made yourself the reason the pour is late.

Here is the part that makes it difficult rather than merely uncomfortable. Most of the time, if you had not stopped it, nothing would have happened. The odds on any single exposure are usually in your favor. That is exactly why the hazard felt tolerable to the crew in the first place. They have done it this way before and walked away every time.

So you are absorbing a certain, visible cost to prevent an uncertain, invisible one. That is a bad trade emotionally and a good trade mathematically, and people are not built to feel the second one.

The test that makes it simpler

The question that cuts through the hesitation is not whether something feels unsafe. Feelings are contestable, and you will lose that argument with a foreman who has thirty years on you.

The question is whether a high-energy hazard is present with no direct control on it. High energy means energy that can kill on contact: a fall above about four feet, a suspended load, an energized system, an unprotected trench wall, equipment in motion. A direct control physically blocks that energy even when somebody makes a mistake. An indirect control depends on every person behaving correctly for the whole duration of the task.

If there is high energy present and the only thing standing between it and a worker is a human being not making an error, that is the stop. It is not a judgment about whether the crew is careful. It is an observation about whether care is the only thing holding the outcome together.

That framing also changes the conversation, because you are no longer telling a foreman his people are careless. You are pointing at a control gap, which is a problem you can both look at.

Raising it without wrecking the relationship

Go to the person doing the work first, not to their boss. Walking past a crew to report them up the chain guarantees you will never be told about anything again, and most of what keeps a site safe is what people tell you voluntarily.

Be specific about the hazard and the missing control rather than about behavior. There is a real difference between telling somebody they are working unsafely and telling them there is nothing between them and a twenty-foot fall except their footing. The first is about them and invites defense. The second is about the situation and invites a fix.

Bring the fix if you have one. A stop with a path back to working in ten minutes is a completely different event than one that ends the shift. Often the correction is available on site and nobody had worked out how to get it in place.

Then, once the hazard is controlled, get out of the way and let them work. The point was never to demonstrate authority.

Documenting it so it holds up

Document the stop even when it resolves in fifteen minutes and everybody parts on good terms. Not to build a case against anyone. An undocumented stop does not exist as far as your program is concerned, and it teaches you nothing.

What the hazard was, described in terms of the energy present and the control that was missing.
The time work stopped and the time it resumed.
Who was notified and what was decided.
What control was in place before work restarted.
Whether the same gap could exist somewhere else on the site right now.

That last item is the one people skip and the one that matters most. A missing trench box on one excavation is a task problem. A crew that did not know a trench box was required is a site problem, and you just found it cheaply.

When you get overruled

Sometimes you stop work and somebody restarts it over your objection. How you handle that determines whether you have a program or just a policy.

Put your position in writing the same day, in a form that leaves a record. An email to the person who made the call, stating the hazard, the missing control, and what you recommended. Not as a threat. As documentation that the decision was made with the information available at the time.

If a high-energy hazard with no direct control is being worked over your objection, that is a genuine escalation. It is worth remembering that you are the person who will have to live with knowing. Most people who have been in this industry a long time can name a moment they let something go, and that memory is what the escalation is protecting you from.

Key Takeaways

Stop work authority fails in practice because the cost is immediate and social while the benefit is an incident nobody sees.
The usable test is not whether something feels unsafe. It is whether a high-energy hazard has a direct control or only a person not making a mistake.
Go to the crew doing the work first, and frame it as a control gap rather than careless behavior.
Bring the fix when you have one. A ten-minute stop lands very differently than an open-ended one.
Document every stop, including whether the same gap exists elsewhere on site.
If you are overruled on a high-energy hazard, put your position in writing the same day.

Building a stop work culture that people actually use is harder than writing the policy, and it is easier with an outside set of eyes. If that is where your site is, reach out to TriCore and let's talk about it.