Back to resources
Updated on September 22, 20268 min read

Virtual queues for video service: how to organize your operation

A virtual queue organizes people waiting for video service. To work well, it needs an entry rule, available staff and a clear alternative when capacity runs out. This guide helps turn those decisions into an operating routine.

The key decision

Start with the service and the team's capacity. Configure the queue only after that: increasing the waiting limit does not create more appointments or more agents.

1. Choose between queue and appointments by service type

List the most frequent contact reasons and what agents need to resolve each. A customer wanting to show an equipment problem may benefit from video. A request requiring prior document review may need preparation before the conversation.

Criteria for choosing the entry flow

Walk-in demand with a dedicated team

Workflow to evaluate
Virtual queue
Required condition
Published opening hours and capacity to handle arrivals.

Specialist or advance preparation

Workflow to evaluate
Scheduling
Required condition
Reserved appointment slots and instructions sent before the meeting.

Variable demand with customer preference

Workflow to evaluate
Queue and appointments
Required condition
Separate capacity and an appointment option that is actually available.

Simple request that does not need a live conversation

Workflow to evaluate
Asynchronous channel
Required condition
Clear guidance and an owner for follow-up.

When the same person handles the queue and scheduled appointments, reserve time for both. Counting the same agent as available in both channels creates capacity that exists only in a spreadsheet. Also record how customers continue if they cannot use a camera or microphone.

2. Write the rules before configuring the tool

A useful rule addresses an observable situation. “Serve quickly” does not explain what to do when the queue fills. Use a short record per service or branch, with an owner and review date.

  • Entry: which topics are accepted, during which hours and what information is needed? Ask only for what will be used during service.
  • Order and priority: what is the default rule and which exceptions are approved? Explain who may change priority and record the reason.
  • Distribution: who may receive calls for this service? Who covers absences? Agent status should reflect real availability.
  • Limit: how many people may wait and who monitors waiting? The limit should consider the experience offered as well as technical capacity.
  • Exit: what message appears outside business hours, when full or on failure? State the next action and a functioning channel.

Skill-based routing, round-robin assignment and lowest-load distribution are different methods. Choose based on service needs and verified platform options. A list of service techniques does not mean all are configurable product rules.

3. Estimate capacity and observe peak-demand periods

Start with a simple calculation to spot mismatches. Divide the team's effectively available minutes by work time per case, including the conversation and subsequent recordkeeping. This provides a capacity reference, not a waiting-time forecast.

Break data down into 15- or 30-minute intervals, by service and weekday. Include breaks, absences, training and off-call tasks. Also examine duration variability: an average can hide a few very long cases.

When demand exceeds capacity, evaluate one change at a time: adjusting shifts, reducing misrouting, reserving appointments or improving pre-service instructions. Compare similar periods before concluding that a change worked.

4. Separate management methods from NEFT features

In NEFT, queue settings include hours, capacity and messages for situations such as a full queue or closed hours. There are company settings and, where applicable, branch-level adjustments. Check the effective value for the branch before announcing hours to customers.

When enabled, automatic assignment considers eligible, available agents. Among them, it distributes calls based on the number of assignments that day and respects branch membership when the interaction identifies one. This does not automatically select someone by language, product knowledge or expected case duration.

The product also has a VIP list. Treat its use as an explicit operational decision: define criteria, review entries and observe the effect on those remaining in the regular queue. Avoid using priorities as a substitute for insufficient staffing.

When enabled and configured, appointments can serve as a queue alternative. Verify this option in the customer's actual journey. Explore the service queue and online appointments pages to discuss your scenario with the NEFT team.

5. Use metrics to choose an action

Define measured events before comparing reports. Does waiting start on the link click, after registration or when camera and microphone are ready? Does duration include subsequent recordkeeping? Mixing definitions creates differences that look like improvement but are merely calculation changes.

  • Waiting: track the average and distribution. If individual data is available, also examine the 90th percentile: the time within which 90% of measured waits ended.
  • Abandonment: divide departures before service by valid queue entries. Define how to handle reconnections and repeated attempts.
  • Routing: record cases that reached the wrong team. They indicate a need to review entry or training.
  • Resolution: track whether the request was resolved and whether there was another contact for the same reason. Ending a call does not prove resolution.

These are definitions for your measurement plan; check the definition of each metric displayed by the tool. Each week, record an observation, chosen action, owner and next comparison date. The ROI guide explains how to connect this data to cost per resolution.

6. Checklist for opening the queue with a clear alternative

  1. An owner approved the services, hours and scheduled staff.
  2. The link was tested on computer and mobile, including camera and microphone permissions.
  3. Available, busy and paused agents were tested in the intended workflow.
  4. Full queues, closing and staff unavailability have understandable guidance.
  5. The alternative offered to customers was tested and has an owner.
  6. Metric definitions and the first review date are recorded.

Validate this first in a test environment or controlled pilot. Increase volume when the team demonstrates it can receive, serve and route exceptions under the agreed conditions.

Frequently asked questions

Discuss your operation with NEFT

Tell us which services, volumes and hours your team needs to cover. Assess the right configuration for your operation.

Find your solution