“Someone watches the cameras” can describe several operating models. The choice starts with what needs noticing and what happens when the scene falls outside an alert rule.
The idea in view · Planning diagram
Two different ways to assign attention
- Defined event
- Associated video
- Triggered review
- Defined viewing period
- Agreed live views
- Scheduled attention
Make the call
Match the review model to the task
| Situation | Decision to make | What to ask for |
|---|---|---|
| A closed gate has a defined entry rule | Whether event selection covers the task | A qualifying event and missed-event test |
| A busy operation needs judgment beyond predefined alerts | Whether dedicated viewing is required | Coverage position, breaks and simultaneous-event handling |
| An incident affects several sites | Who prioritizes and maintains ownership | A workload and escalation discussion |
Identify what needs attention
A closed yard with defined entry rules may be suitable for alert driven review. A busy operation that needs observation beyond predefined events may need a dedicated viewing position. Ask the provider to describe what happens between alerts and whether a human is actually watching then.
Test the selection boundary
With event based review, test what creates an event, what gets filtered and what happens when a clip is unavailable. With live viewing, ask about camera rotation, breaks, simultaneous incidents and handovers. Neither model makes an obstructed view useful.
Compare the actual service
Write down viewing hours, included actions and backup arrangements. NightMarshal describes AI checks with human exception review; do not infer a dedicated person continuously watching every camera. Dedicated staffing and event based review should be evaluated as different scopes.
Ask what happens between events
In an event-driven model, the important question is what brings an event to a person. Ask about the rule, any filtering, the available clip and the path when a device stops reporting. If activity never becomes an event, the reviewer may have nothing to review. A demonstration should include that selection boundary, not just the operator opening a clean test clip.
In continuous live viewing, examine the viewing duty itself. How many agreed views are in the position’s scope? What happens during breaks, maintenance or two incidents at once? Provider labels do not answer these questions consistently. SCW, for example, describes alert-triggered human review, while MonitorCrew describes separately scoped staffed coverage. Ask for your proposed model in writing.
What does event-based monitoring actually do?
Event-based monitoring uses a defined trigger to decide when review work begins. The trigger might come from camera analytics, another alarm system or an approved integration. The reviewing service needs enough context to assess that event and follow the agreed response plan. Recording can continue independently of review, so a system that records continuously is not necessarily a system whose live views receive continuous human attention.
The details matter more than the label. Ask which events are sent, what video accompanies them, which hours apply and what happens if event delivery fails. An event-based service may suit a task with clear conditions, but its usefulness depends on the trigger and the review path as well as the camera view. Events outside the accepted rules may not reach the reviewer even if a local recorder retains footage of them.
What does continuous viewing include?
Continuous viewing assigns ongoing attention to agreed live views during specified periods. It still needs a defined scope: which views, how they are presented, what duties the reviewer performs and what happens during interruptions. A promise that cameras are watched does not explain whether every view is displayed continuously, rotated, supported by events or reviewed only when a concern is raised. Ask for the actual operating model.
Some tasks benefit from context across a period of activity, such as a busy site opening or a controlled operation. Others may be better expressed as defined exceptions. Neither model gives a reviewer an unobstructed view or the authority to intervene physically. Both need lighting, connectivity, clear normal-activity information and an escalation plan. Compare the duties against the task rather than assuming that one model is universally more protective.
Where hybrid arrangements need careful definition
A service can use events to draw attention while also assigning scheduled live review. That may be a reasonable combination, but it should not be described vaguely as both models being included everywhere. Write down the scheduled periods, event conditions and which path takes priority. The business should know what happens in the intervals between scheduled duties and when the event system is unavailable.
Ask how the provider records each kind of work. An event log can show that a notification was reviewed; it does not demonstrate uninterrupted viewing of every camera. A scheduled viewing record may establish that a duty was performed without showing whether every possible event was noticed. Select evidence that matches the accepted task. If the provider cannot explain the distinction, the operating scope needs more work before price comparison is useful.
A fictional choice for a warehouse and its yard
Consider a warehouse whose equipment cage should be empty after closing, while late deliveries are common in the main lane. A team might assess event-based review for an agreed cage boundary and a separate viewing duty for a scheduled delivery operation. That is a planning example, not a recommendation that either model is suitable without testing. The lighting, event rules, view obstruction and local response still require acceptance.
The team should ask what changes when a delivery is early, the cage camera is obscured or the connection drops. If the answers rely on a site contact, that contact needs an agreed role and fallback. If there is no accepted way to review a particular condition, the gap should stay visible. A sensible choice can be a narrower, well-defined duty rather than an expansive label that nobody can demonstrate in operation.
Before you move on
- List tasks that cannot be described by an event rule.
- Ask who watches which views and at what times.
- Test alert or staffing gaps before making a commitment.
A worksheet you can actually use
Start with the blank sheet. Add one row per item, assign an owner and mark anything unresolved as pending. The completed example shows the level of detail to aim for.
Keep completed security plans private. Use aliases; keep passwords, access codes, personal contacts and footage links in your approved systems.
How to fill the worksheet
- Task
- State the operating question the business needs supported. Avoid starting with a service label before defining the duty.
- Event rule feasible
- Record whether a supported trigger can represent the task, with the relevant test or a clearly identified pending question.
- Viewing hours
- Specify periods of assigned live viewing separately from continuous recording or the availability of an application.
- Human coverage model
- Describe who reviews what and when, including the accepted relationship between event-triggered and scheduled work.
- Gap handling
- Identify how an unavailable view, missed trigger or absent contact is handled and who accepts the unresolved condition.
A worked example
Fictional buyer: an overnight gate rule fits alert review, while a scheduled loading operation requires an additional task discussion. The team obtains separate scopes instead of assuming one covers both.
| Task | Event rule feasible | Viewing hours | Human coverage model | Gap handling |
|---|---|---|---|---|
| Closed gate access | Yes subject to testing | Overnight | Exception review | Fault and backup plan |
Questions readers bring to this task
What else should you know?
Can a site use both approaches?
Yes, when the duties and handoff are clear. A site might need a dedicated viewing task for one operating period and event review when closed. Quote and test each part; do not assume that adding cameras automatically adds continuous human observation.
Link to this answer ↗Does continuous recording mean someone watches continuously?
No. Recording describes what the system stores; monitoring describes the review duty. Ask separately about recording schedules, event delivery, live viewing periods and human review. Keep each answer in the scope so a storage setting cannot be mistaken for a staffed service.
Link to this answer ↗Is event-based monitoring always cheaper?
No universal comparison follows from the label. Cost depends on the accepted views, event workload, hours, implementation and response duties. Compare written quotes for the same operational requirement and identify what each model includes or excludes before drawing a cost conclusion.
Link to this answer ↗What happens when an event is missed?
The team needs to determine whether the issue was visibility, detection, event delivery or review. Local footage may help reconstruct the path, but recording the event does not prove the monitoring service received it. Preserve the relevant evidence and use the missed-event guide to investigate the actual boundary.
Link to this answer ↗Notes & sources
Provider terms differ. This comparison describes operating models, not a hands-on product test.
Reviewed October 27, 2026. The examples and worksheets are planning aids, not customer results or a statement of contracted service.
- SCW: professional camera monitoring
Provider description of alert review and scoped response. Its offer is not a NightMarshal commitment.
- MonitorCrew: service scope and pricing factors
Provider description of staffing, hours, handovers and scope. No comparative testing is claimed.