Lone worker safety, and what an employer is actually on the hook for
By Jim Hankins
A remote worker and a lone worker are not the same person, and confusing them is how organizations end up buying the wrong thing.
A remote worker is at a desk somewhere that is not your office. If something goes wrong, they pick up a phone. A lone worker is physically alone in a place where something can go wrong and nobody would know for hours. A housekeeper on the fourth floor of a hotel. A janitor in a building that emptied at six. A home health aide at an address nobody else has visited. A technician at an unstaffed site forty minutes down a service road.
The distinction matters because it changes what you owe them. A remote worker needs connectivity. A lone worker needs somebody to notice.
The duty is the noticing
Strip away the vendor language and almost every lone worker obligation, whether it comes from a statute, an insurer, a client contract or your own conscience, reduces to three things:
Somebody knows where they are supposed to be, and when.
They can raise an alarm without a struggle.
Somebody finds out if contact is lost.
Read that list again and notice what is missing. None of it requires a sensor. None of it requires a call center. Two of the three are about a schedule and a person who is accountable, which is why the organizations that do this well are often not the ones who spent the most.
Where it goes wrong is almost always the third one. The first two are easy to buy. A device that sends an alert is a solved problem, and there are a dozen vendors who will sell you one. What is hard, and what most purchases quietly skip, is guaranteeing that an unanswered alert does not simply sit there.
A notification is not an escalation
This is the single most useful question to ask any vendor in this category, including us:
When the alert goes out and nobody responds, what happens next?
If the answer is "it notifies the supervisor", you have bought a notification. A notification is a one-shot event aimed at a person who may be driving, asleep, off shift, or out of signal. It discharges the system's obligation and nothing else. The worker is still alone.
An escalation is different in kind. It keeps going. It moves down an ordered list of people who agreed in advance to be reachable, and it does not stop because the first attempt failed. It stops when a human being acknowledges.
That difference is the whole of lone worker safety, and it is why the ordered list has a name. An alert roster is the long-standing emergency services and military term for exactly that: the sequence of people a call-out reaches, in order, until somebody answers. It predates all of this software by decades.
What a check-in is, in practice
A check-in is a commitment with a deadline. A worker says, in effect, "I will confirm I am fine by 4:30." If 4:30 arrives and the confirmation does not, the system treats the silence as the signal and starts working the roster.
Three things about that are worth sitting with.
Silence is the trigger, which means the worker does not have to be able to act. A device that requires a button press assumes the person is conscious, has a free hand, and is not being prevented. A deadline assumes none of that. It is the only mechanism in this category that works when the worker can do nothing at all.
The check-in has exactly one owner: the person who armed it. Not their employer, not a dispatcher, not us. They set it, they clear it, they cancel it. This is not a philosophical preference, it is what keeps the thing from becoming surveillance, and workers can tell the difference immediately.
It produces a record. Who was on, who was called, in what order, who acknowledged and when. For anyone whose obligation came from an insurer, a client contract or a regulator, the record is frequently the actual deliverable. Nobody audits the intention.
Where we are honest about the limits
Three limits, stated here rather than discovered later:
Nobody at our company is watching. There is no staffed center. Alerts go to the roster you configure, and escalate through it until one of your people acknowledges. Your people, not strangers. That is also a large part of why the price is what it is.
We contact an emergency service in exactly one case. If your organization turns on automatic SOS for a Garmin inReach registered to a worker, and they miss a check-in while their phone has no signal, the app declares an SOS on that inReach and Garmin Response, Garmin's emergency coordination service, receives it under your organization’s Garmin plan. Outside that, we contact nobody but your own roster. If a prospect assumes otherwise in either direction, we correct it on the spot, because a false belief about who is coming is worse than no system.
We are not a cheaper version of a monitored device service. If what you need is a certified alarm receiving center with a police response path, that is a different category and you should buy from it.
The honesty about limits is the asset. An employer who believes we do something we do not, and finds out after an incident, is a lawsuit and the end of a working relationship. We would rather lose the deal in the first conversation.
Where to start if you are responsible for people who work alone
Do this before you look at a single product.
Write down who is actually alone, and when. Not job titles, shifts. The answer is usually smaller and more specific than people expect.
Name who is accountable if one of them does not come back. If nobody in your organization owns that outcome, no product will fix it.
Build the roster before you buy the tool. Who gets called, in what order, and who is the backstop at 2am on a Sunday. If you cannot fill in the second and third positions, that is the finding, and it is more valuable than anything a vendor will tell you.
Then ask every vendor the escalation question above, and listen for whether the answer stops at "notifies".
If a specific obligation is driving this, a state rule, a client requirement or an insurer's questionnaire, that narrows it fast, because the text usually tells you precisely what the device has to do. We have written up the ones we have read against primary sources.
Work out what your people actually need
If you have staff who finish a shift alone and nobody would know for hours, that is a specific problem with a specific answer. It usually takes one conversation to tell whether we are it.
Book a call — 20 minutes, and you will leave knowing whether this fits, including if the answer is no.
What AlertRoster is, and is not
AlertRoster is a call-out notification and escalation tool. It notifies people who have agreed in advance to be notified, and records what happened. It is not an emergency service. It contacts one on your behalf only when your organization has turned on automatic SOS for a Garmin inReach registered to you: if you miss a check-in while your phone has no signal, the AlertRoster app declares an SOS on that inReach, and Garmin Response, Garmin's emergency coordination service, receives it under your organization's Garmin plan. It is not a fire alarm, a security alarm, or an alarm monitoring service, and it holds no life-safety certification. It is not a replacement for your team's official paging arrangements, and it should not be the only way a call-out can reach your people.
Sources reviewed October 2026. Statutes and rules change; if you are reading this long after that date, check the current text.
About the author
Jim Hankins is the founder and CEO of Cloud Bedrock, LLC and the developer of AlertRoster, which he first launched in 2009. A US Army combat veteran and a 42-year veteran of IT and software development, he has spent much of his career building emergency communications platforms, including as CTO of GEOS, the international emergency response coordination service later acquired by Garmin.