How does a queue management system work?
Most systems follow the same five steps.
- Join. A person takes a numbered ticket from a machine or a desk, books a place on a phone or kiosk, or simply joins one line. In a virtual queue the person checks in and gets in line remotely, so they hold a place without standing in it. [3]
- Wait and be informed. A display or a message shows which number is being served. Some services also publish their current waiting times so people can plan their visit. [2]
- Be called. The next person is called to a free counter by number, screen, sound or a message to their phone.
- Be served. The counter records when the service starts and ends.
- Measure. The system reports waiting time, service time and volumes by hour, service and counter.
The first four steps organise the queue. The fifth measures it, and the measurement is only as complete as the data behind it.
What is the difference between ticket waiting time and real waiting time?
Ticket waiting time runs from the moment a ticket is issued to the moment it is called. Real waiting time runs from the moment a person walks in to the moment they reach the counter. The two differ whenever people wait before taking a ticket, stand in a line that has no ticket, take a ticket and leave, or come back later for the same matter.
Public services draw the same line. California's Department of Motor Vehicles, for example, has reported the time a customer waits in line before being issued a queue ticket, which it calls the pre-queue, separately from the time after checking in and receiving a queue number. [2]
A system that measures only tickets cannot see the wait before the ticket, so a centre can report short ticket waits while the hall still feels full. Measuring the real wait means observing the floor itself, by zone and by counter, not only the ticket log.
How are queue length and waiting time related?
They are tied together by a well-known result from queueing theory, Little's law. For a queue in a steady state, the average number of people in the system (L) equals the average rate at which they arrive (λ) multiplied by the average time each one spends in the system (W): L = λW. John D. C. Little published the proof in the journal Operations Research in 1961. [1]
In practice, if arrivals hold steady and the average time in the system doubles, the average number of people in it doubles with it. That is why a queue system that reports both the length of the line and the wait gives a more honest picture than either number alone.
What should you check before choosing a queue management system?
A buyer checklist, in plain questions.
- What does it measure: tickets, or people? Ask whether waiting time starts at the ticket or at the door.
- Does it see every counter on its own? Busy and idle counters, and how long each service really takes at each one.
- Can it alert before the service level is breached? On which channels, and who acknowledges the alert.
- Can it tell a returning visit from a new one? A person who comes back for the same matter is a different problem from a new visit.
- Where is the data processed and stored? On the device, in your own cloud tenant, or on your own servers, and whether any images or video leave the site.
- Does it work with what you already run? Your ticketing system, displays, CRM and existing CCTV, scoped with your team rather than assumed.
- What does the team actually see? The dashboards, the reports and the actions they lead to.
- How is it sized and deployed? Ask for a site survey, not a rule of thumb.
How do you compare queue management options?
Vendor-neutral criteria for a shortlist.
| Criterion | What to ask | Why it matters |
|---|---|---|
| Scope | Does it organise the queue, measure it, or both? | A site can keep its ticketing system and add measurement alongside it. |
| Start of the wait | Ticket issue or entry to the hall? | Decides whether the reported wait matches what people experience. |
| Granularity | Per zone, per counter, per service, per hour? | Averages hide the counter or hour where waits build. |
| Alerting | Thresholds, channels, acknowledgement trail | An alert nobody acknowledges changes nothing. |
| Repeat visits | Can it link a second visit to the first, without storing identity? | Second visits for the same matter point at a service problem, not a staffing one. |
| Data handling | Processing location, what leaves the site, hosting options | Some buyers need data to stay on site or in-country. |
| Integration | Ticketing, CRM, displays, existing CCTV | Avoids running two systems that disagree. |
| Deployment | Site survey, pilot scope, who installs and maintains | A survey of your own floor plan is the only reliable basis for a design. |
Where does Pulses fit?
Pulses.ai is not a ticket system and does not replace yours. It adds the measurement a ticket log cannot give: each customer's visit seen from the floor, from the door to the counter. A real customer-by-customer journey.
- The real wait. Pulses Vision devices on the ceiling measure each customer's visit from the door to the counter, and show where they waited on the way. See Queues and waiting times in service centres.
- Each counter on its own. Pulses Ops shows each zone and counter, busy or idle, and how long a service really takes at each one.
- Told before the service level breaks. When a wait nears your target, the supervisor hears on WhatsApp, Teams, SMS, email or an automated voice call, the channels the team already answers. See Branch queues, waits and repeat visits.
- Unique visitors and second visits. Pulses CX shows the same customer back for the same matter, even at another branch, as one journey. This works through Anonymous Unique FaceID: derived on the device, it can't be turned back into a picture and holds no personal data.
- Mood tied to the wait. Pulses CX reads mood per zone with no surveys and ties it to the wait that caused it.
- Processed on the device. No image leaves the device; only anonymous metadata does. See Edge Intelligence.
- Your hosting choice. Pulses runs on Pulses Cloud, in your own Azure tenant, or fully on-premise on the Pulses Appliance.
The same approach applies wherever people wait: airport check-in and security, café counters and drive-thrus, and government service centres.
