Lone Worker App Overview

Lone Worker

1. What it is

Lone Worker is a safety check-in and SOS escalation app for frontline workers. Workers confirm they are safe on a timed cadence; if they miss a check-in or tap the SOS button, an alert climbs an ordered responder ladder — push, SMS, and voice — until a human acknowledges it.

• Enablement: Lone Worker is a licensed, admin-enabled app. An admin turns it on from Admin → Apps → All Apps and configures at least one responder roster before check-ins arm. An SOS raised with no roster still pages your safety admins as a break-glass fallback.
• What it is not: Lone Worker does not track location continuously. It captures a single GPS reading at the moment an SOS is triggered, if the admin has left that setting on.

2. Standing it up

1. Go to Admin → Apps → All Apps, find Lone Worker, and enable it for your business.
2. Go to Lone Worker → Responder Rosters and create at least one roster. A roster is scoped to a site (location) or a team (scheduling group). Choose the scope from the grouped picker in the new-roster drawer.
3. On the roster detail page, add at least one tier. Each tier needs at least one responder and a set of notification channels (push, SMS, voice). A roster with no populated tier has nobody to notify.
4. Check-in coverage starts from a roster, not the app toggle. A worker’s shift must resolve to an active roster before check-ins arm. Enabling the app without creating a roster does nothing — no workers will be prompted.
5. Optionally, go to Check-in Schedules and create a schedule for a site or team to override the default check-in interval and grace window.
6. Go to Settings to adjust defaults (check-in interval, grace window, escalation timeout) and to choose which optional features run. Location capture and the Safety Hub incident handoff are on by default; Live Assist video and wearable alerts are off until you turn them on.
7. To confirm it works: clock a worker in on a shift at a site that has a roster. Within one check-in interval the worker will see an “I’m safe” prompt on their My Safety page and receive an Inbox notification with push. The Dashboard will show the active roster count and any live alerts.

Silent-until-paired warning: A roster scoped to a site only covers shifts at that site. A worker on a shift at a different site — or on a shift with no site at all — has no roster and will never receive a check-in. If you expected coverage and see none, verify the roster’s scope matches the shifts you are running.

Who sees what: The app sidebar has two groups. Every user with the app enabled sees My Safety and Alerts in the Safety group. Dashboard is there too for business admins, Lone Worker app administrators, anyone on a responder roster, and anyone either permission setting below covers; for everyone else, opening Lone Worker goes straight to My Safety. The Manage group shows only what the viewer is allowed to open: Responder Rosters and Check-in Schedules for people covered by the Who can manage responder rosters & check-in schedules? setting, Analytics for people covered by Who can view safety analytics?, and Settings for business admins and Lone Worker app administrators. Both permission settings default to admins and app administrators only.

3. How it fits together

Responder Roster — the ordered list of people to contact when something goes wrong. Each roster is scoped to a site or a team. The most specific active roster wins: a team roster overrides a site roster. If no roster resolves, the alert routes to your safety admins as a break-glass fallback.

A roster has one or more tiers, each with its own responders, notification channels (push, SMS, voice), and an acknowledgement timeout. The escalation engine walks the tiers in order: it notifies the first tier, waits for the timeout, and if nobody acknowledges, climbs to the next. A tier marked Break-glass also notifies your safety admins on top of its own responders. Safety admins are every business admin unless you pick a group under Settings → Escalation → Safety admins (only group members who are business admins or Lone Worker app admins are paged, since only they can open the alert and respond to it; a group with nobody like that in it falls back to every business admin, so the page is never lost).

Worked example: A warehouse has a roster named “Warehouse Night” scoped to the night-shift site. Tier 1 is the floor supervisor (push + SMS, 120-second timeout). Tier 2 is the safety manager and the site lead (push + SMS + voice, 120-second timeout). Tier 3 is break-glass. A worker triggers an SOS → the floor supervisor’s phone buzzes. Two minutes pass with no acknowledgement → the safety manager and site lead receive push, SMS, and a voice call. Two more minutes → the break-glass tier fires: the safety admins are notified across all three channels. The climb stops the instant anyone acknowledges — for 60 minutes. If no outcome is recorded in that time, the alert goes back to the tier it was on and pages it again.

Check-in — a timed “I’m safe?” prompt armed automatically when a worker clocks in on a covered shift (a shift whose site or team resolves to an active roster). The interval and grace window come from a check-in schedule for the site or team, or from the app-wide defaults if no schedule exists. When a worker answers, the next check-in is re-armed immediately. When the grace window lapses without an answer, a missed check-in alert fires through the same roster escalation as an SOS. Check-ins stop at the shift’s scheduled end, even if the worker is still clocked in, and the worker is sent a notice saying so; their SOS keeps working.

Check-ins are armed and cancelled by the clock-in/out hook and by a background sweep that runs every five minutes. The background sweep also sends the worker an Inbox/push notification when their check-in is coming due, so they do not have to keep the My Safety page open.

Check-in Schedule — an optional per-site or per-team override of the default check-in interval and grace window. Only one schedule per site or team is allowed. When no schedule covers a shift, the app-wide defaults apply. Schedules are managed under Lone Worker → Check-in Schedules.

Alert — the escalation record. Alerts are raised by three triggers:

• SOS — one-tap emergency from the worker’s My Safety page or the mobile app. Only one active SOS per worker at a time; a second tap while one is live does not restart the ladder.
• Missed check-in — raised automatically when a check-in’s grace window closes without an answer.
• Floor issue — a worker-reported equipment or line problem (“Report a floor issue” on My Safety). Follows the same roster escalation but does not capture GPS.

An alert ends in one of two terminal states: resolved (safe, false alarm, or incident) or exhausted. An alert is exhausted when the responder ladder runs out with no outcome recorded — nobody acknowledged in time, an acknowledgement lapsed and nobody took it again, or the responder who acknowledged it handed it back (or left the roster) with nobody on that tier or above left to alert — and that fires the dead-man’s switch. One more case ends as exhausted without paging anyone: an acknowledgement still open after 24 hours is closed automatically. An exhausted alert can still be resolved: open it and record what happened.

Dead-man’s switch — if the entire ladder runs out with no outcome recorded, your safety admins are notified in the app and by push, email, SMS, and voice call. No Lone Worker setting turns it off. The in-app, push and email notices follow each admin’s own notification settings for Safety Hub & Lone Worker; the text message (to a verified mobile number) and the phone call go out whatever those settings say, as long as texts and calls are on for your organization.

Safety Hub handoff — when a responder resolves an alert as Incident, Lone Worker can open a Safety Hub incident automatically (on by default; controlled by a setting). The incident inherits the alert’s location, shift, and worker. A floor-issue alert that resolves as an incident is categorised as an equipment failure; all other triggers are categorised as security.

My Safety — the worker-facing surface at Lone Worker → My Safety. Workers see:

• Their active check-in prompt with an “I’m safe” button (always visible while a check-in is armed, not only when due).
• A one-tap SOS button.
• A Report a floor issue button for equipment or line problems. A short note describing the problem is required.
• Their recent alert history (last five alerts).
• Their currently active alert, if any.

On mobile devices this page redirects to the mobile-optimised view at /m/apps/lone-worker.

Audit trail — every alert records an append-only event log: raised, each tier notified (with per-channel delivery outcome), acknowledged, resolved, and exhausted. These events are never modified or deleted. They are visible on the alert detail page and exportable as CSV for EHS or regulator review.

4. Running it

Reviewing the dashboard

Go to Lone Worker → Dashboard. The dashboard shows four counts: Open alerts (raised, notifying, or acknowledged), Resolved today, Escalation exhausted (exhausted alerts with no outcome recorded yet), and Active rosters. Below the counts is the Open alerts table — the 25 most recent, with View all open for the rest. When nothing is open, an admin sees No open alerts — everyone’s accounted for (or, while no real roster exists yet, a prompt to create one). If exhausted alerts still have no outcome, it says No alert is climbing the ladder — but exhausted alerts still need an outcome instead, with a link to them.

Admins and app admins see all alerts across the business. Other users see only their own alerts and alerts on rosters where they are listed as a responder.

Responding to an alert

When notified (push, SMS, or voice), open the deep link to the alert detail page. The alert shows the worker’s name, the trigger type, any note the worker attached, the location (if captured), and the full event timeline.

Select I’ve got this — acknowledge to acknowledge the alert and take ownership. Everyone else who was paged is told to stand down, and the climb holds for 60 minutes. Record the outcome within that time, or the alert goes back to its tier and pages it again. If the Open a live video channel to the responder when an alert is acknowledged setting is on, a Live Assist video session opens automatically so the responder can see the worker’s camera. It applies to every alert that is acknowledged, not only an SOS.

Then resolve the alert with an outcome:

• Safe — the worker is confirmed safe; the alert closes.
• False alarm — accidental trigger; the alert closes.
• Incident — a real safety event occurred. A note is required. If the Safety Hub handoff setting is on, a Safety Hub incident is created automatically.

A worker cannot acknowledge their own alert — that would stop the climb with nobody coming. A worker who is not a responder can close their own alert only as False alarm. A worker who is also a responder on the alert’s roster, a business admin, or a Lone Worker app admin can record any outcome on their own alert, like any responder — they still cannot acknowledge it.

Filtering and exporting alerts

Go to Lone Worker → Alerts. The alert index supports filtering by state (Open, Active, Acknowledged, Resolved, Exhausted — Open is raised, notifying, or acknowledged), trigger type, date range, and individual worker. The same filters apply to the CSV export.

Admins, app admins, and anyone on a responder roster can export a CSV of the filtered alert list. Each row records the alert id, trigger type, any trigger note, worker, state, resolution, raised/acknowledged/resolved timestamps, and linked Safety Hub incident id.

The export is scoped to the viewer’s own purview — a non-admin responder’s CSV contains only alerts they could already see on the index page.

Managing rosters and tiers

Go to Lone Worker → Responder Rosters. From the roster detail page you can add and remove tiers, add and remove responders on each tier, set per-tier notification channels and acknowledgement timeouts, and mark a tier as break-glass. Edit renames a roster or changes the site or team it covers, and Delete removes it along with its tiers. There is no switch to pause a roster.

When adding a responder to a tier, each person can appear on that tier only once. A responder who is also a business admin will not be double-notified on a break-glass tier.

Viewing analytics

Go to Lone Worker → Analytics (visible to the people the Who can view safety analytics? setting allows — admins and app administrators by default). Analytics shows business-wide totals only, with no worker names, locations or notes. The page shows total alerts, acknowledged rate, average acknowledgement latency, exhausted-but-open count, check-in completion rate, and a 14-day alert volume chart. An analytics CSV export is also available.

Loading sample data

Go to Lone Worker → Settings and select Load Sample Data to populate responder rosters, check-in schedules, check-ins, and alerts with realistic demo data — recent activity plus alert history going back several months. Loading runs in the background and can take a few minutes. Sample data never fires real notifications and can be taken out with Remove Sample Data, which also removes any Safety Hub incident created by resolving a sample alert as an incident.

5. Settings

All settings are at Lone Worker → Settings (admin or app admin only).

Setting	Default	What it changes
Enable the Lone Worker AI Agent in Ask AI	On	Whether the Lone Worker agent answers questions in Ask AI and can acknowledge an alert for you, or hand back your acknowledgement (it always asks first).
Who can manage responder rosters & check-in schedules?	Admins and app administrators only	Who can create, edit and delete responder rosters, tiers and responders, and check-in schedules. You can widen it to every manager, or add specific groups. Anyone allowed here can add themselves to a roster, and a roster’s responders see its alerts.
Who can view safety analytics?	Admins and app administrators only	Who can open Analytics and its CSV export. You can widen it to every manager, or add specific groups. Analytics never shows worker names, locations or notes.
Default check-in interval	60 minutes	How often a worker is prompted “I’m safe?” when no site/team schedule overrides.
Default grace window	10 minutes	How long after a check-in is due before a miss fires an alert, when no schedule overrides.
Default acknowledgement timeout per tier	120 seconds	How long a tier waits for an acknowledgement before climbing to the next tier, when the tier does not set its own.
Open a Safety Hub incident when an alert resolves as “Incident”	On	Whether resolving an alert as Incident automatically creates a Safety Hub incident record.
Capture device location when an SOS is triggered	On	Whether the worker’s device GPS is captured once, at trigger time, and attached to the alert.
Open a live video channel to the responder when an alert is acknowledged	Off	Whether acknowledging an alert opens a Live Assist video session between the worker and the responder. Requires the Live Assist app and a configured LiveKit provider.
Let wearable & IoT devices raise alerts	Off	Whether events from a connected wearables integration (badge presses, panic buttons, line sensors) can raise alerts and page responders. An admin connects a device platform under Admin → Integrations → Wearables & IoT (signing secret, fallback worker, and the webhook URL to give the vendor); until it is connected and this is on, device events raise nothing.

6. More help

• Lone Worker FAQ — specific setup, operating, and troubleshooting questions.
• Ask AI — the assistant answers questions about Lone Worker from these articles.