Retail's danger clock

Ask a retail worker when the floor feels different and they will tell you: after dark. The FY26 monitoring data puts numbers on the instinct.

Retail organisations generated 41.8% of all duress activations in the FY26 State of Lone Worker Safety report - the largest share of any sector, which is what you would expect from the sector that puts the most workers face-to-face with the public. What’s more interesting is when those alarms turn out to be real.

The clock

Confirmation rates for retail-sector activations by time of day:

• Mid-morning (6am to 9am): 3.0% - the floor’s safest stretch.
• Evening trade (6pm to 9pm): 7.4% - well over double the morning rate.
• After midnight (12am to 3am): 8.4% - the highest-risk window of all.

The evening block is late trade: skeleton staffing, alcohol in the mix at attached outlets, tills full, and the day’s friction arriving at the counter. The after-midnight block is the smaller set of sites that never close - and when an alarm goes off at 1am, it is real more often than at any other hour.

Days of the week matter less than hours. Retail’s confirmation rate stays in a band across the week rather than spiking on weekends the way other sectors do. Retail risk doesn’t take weekdays off; it clocks on in the evening, every evening.

Retail presses the button

The other retail signature in the data: immediacy. Automatic triggers - safety timers, missed check-ins, fall detection - make up just 3.8% of retail activations, against a cross-sector average of 17.7%. Retail alarms are overwhelmingly a worker pressing something, right now, because the threat is standing in front of them.

That shapes what a retail duress system has to be:

1. Reachable without a phone. A worker mid-confrontation cannot unlock a screen. A duress alarm they can press without looking - a wearable, a fixed button, a card on a lanyard - is the difference between an alarm and a wish.
2. Silent. Escalating an aggressor with a siren makes the situation worse. The alarm’s job is to bring help, not attention.
3. Answered in seconds, at 8pm as much as 10am. The 7-second median operator pickup in our data does not vary by hour, which is the entire point: retail’s risk clock runs into the night, and the response has to run with it.

The full dataset and methodology are in the FY26 report, free to cite with attribution. Sector figures are shown as rates, classified by organisation type, so no individual customer or region is identifiable. More on retail deployments: Duress for retail.