Back to Base Monitoring vs AI CCTV Monitoring in Australia

What is back to base monitoring?

Back to base monitoring is a service where alarm signals travel over a monitored path to a graded control room, staffed around the clock, whose operators verify the signal and action the agreed response plan. In Australia those centres are graded under standards such as AS/NZS 2201.2, and the response usually means calling a keyholder or dispatching a patrol.

The term comes from the signalling arrangement: the alarm panel reports "back to base", meaning back to the monitoring centre, rather than only sounding a siren at the property. It is the oldest and best understood model in Australian security, and for a large share of commercial sites it is still the correct one. This guide sets out how it works, what drives the price, the situations where a graded centre is the right answer, and where AI camera verification fits alongside it.

One compliance point applies throughout. Australian security licensing is set state by state, and what you may lawfully offer depends on the activities you actually perform. Software that filters and verifies camera events does not remove any licence you would otherwise need, and certified dispatch and physical response belong to appropriately licensed parties. Confirm your state or territory requirements before offering a monitoring service. This is general information, not legal advice.

How back to base monitoring works

Four parts make up the service. The panel or camera system on site generates a signal. The signalling path carries it to the centre, historically over a phone line and now most often over IP with a mobile backup, monitored so that a cut path is itself an alarm. The graded control room receives the signal, and a licensed operator applies the response plan. The response is whatever the contract says: a call to the keyholder list, a patrol attendance, or a police call where the signal meets the threshold for one.

The grading is the part that carries the weight. A graded centre commits to redundant power and communications, physical security of the room itself, audited procedures, retained records, and continuous staffing. That is a substantial standing cost, and it is what a customer is buying when a contract specifies graded monitoring.

What drives back to base monitoring costs

Providers quote differently, so compare the structure rather than the headline number. The recurring components are usually a monthly fee per site or per monitored panel, the cost of the signalling path, and any service that sits outside the base plan. The variable component is the one that surprises people: callouts and guard attendance are commonly billed per event.

That makes the false alarm rate at a site a large part of what monitoring ends up costing. Every unnecessary signal consumes operator minutes, and every unnecessary attendance is billed. It is also why upstream filtering has commercial value to both sides of the relationship. Before comparing quotes, ask each provider for the base fee, the per event charges, and the response plan in writing.

When back to base monitoring is the right answer

A graded centre is the correct choice more often than the AI conversation suggests. Reach for it when:

• A contract or tender specifies it. Government, banking, pharmacy, licensed premises and many commercial leases name graded monitoring explicitly. If the document says graded, the answer is a graded centre.
• An insurer requires it. Cover on higher value stock and premises is frequently conditional on monitored alarm coverage and a defined response. Read the policy schedule before changing anything about how a site is monitored.
• The site needs guaranteed human dispatch. Where somebody must physically attend within an agreed time, that commitment comes from a staffed centre with a patrol contract behind it.
• Police response is part of the plan. Where a signal may result in a police call, it needs to come through the channels and verification procedures that police accept.
• The site is a full alarm system, not a camera system. Panels, sensors, duress and open or close reporting are what back to base is built around, and a camera layer does not replace that.

Where AI camera verification fits

AI camera verification is a software layer over cameras that are already installed. Still images captured on motion are analysed, and the software classifies what is in them, separating a person or vehicle from animals, weather, headlights, foliage and shadows. Verified events raise an automated phone call, SMS or email to the contacts you nominate. There are no human operators on Vael's side, which is exactly why it is priced and used as a filtering and notification layer rather than a response service.

It earns its place in three situations:

• Unattended yards and compounds with cameras but no panel. Plant and equipment yards, contractor depots and construction sites often have good camera coverage and nothing watching it overnight. Verification turns that footage into a notification while something can still be done about it.
• Sites where false alarms are the actual cost problem. When a site is generating attendances that turn out to be a possum or a swaying branch, filtering upstream reduces the bill without reducing the coverage. Our guide to the common causes of remote CCTV false alarms covers what triggers them.
• In front of an existing graded centre. Pre-screening can sit upstream of a control room and hand only verified events to the CMS already in use, by SMS or email. The grading, the operators and the response plan are untouched. This is how Vael works with A1 and A2 monitoring stations.

Vael is quoted per site rather than per event, so a noisy month does not produce a surprise invoice. Detection covers three products: Security Monitoring, Fire and Smoke Detection and Gate Watching.

The monitoring models available in Australia

Between a fully graded service and a purely automated one there are several arrangements in common use. Most Australian operators end up combining two of them.

Back to base with a graded alarm receiving centre

You sell and install the system, then route signals to an existing graded centre under a wholesale agreement. The centre provides the licensed operators and the dispatch path; you keep the customer relationship. Suits any site where certified human response is part of what is being sold. The trade-off is that false alarms still reach operators, and the per event charges follow.

Outsourced remote video monitoring

A specialist provider watches your customers' cameras on your behalf, often with live verification and audio deterrence. You resell it under your own brand. Suits high risk sites such as construction, vehicle storage and logistics that need live human eyes. The trade-off is a higher per camera cost and a dependence on the provider's staffing and service levels.

Automated camera verification with notification only

Software analyses every event and raises an automated call, SMS or email to the site contact or keyholder, with no operator in the loop. Suits lower risk commercial sites where notification is genuinely what the customer needs. The trade-off is that automated notification is a different commitment from graded operator dispatch, so it has to be matched to the site's risk and to what the customer's insurer requires. Note that operating an automated monitoring service can itself be a licensable security activity in some states and territories.

Verification in front of a graded centre

AI pre-screens the camera stream of events and passes verified ones to a graded centre for action. Suits stations carrying heavy false alarm volume, and installers who want a graded response on the end of the chain without the noise. The trade-off is an integration step at the CMS.

Self-monitoring with filtered alerts

The business watches its own cameras, with software filtering what reaches the phone. Suits small sites and tight budgets. The trade-off is that nothing is guaranteed to happen unless the owner acts, and alert fatigue is a real failure mode when filtering is poor.

Scheduled split between AI and graded response

Automated verification during occupied hours, escalating to a graded centre or patrol overnight when the site is empty. Suits sites with clear occupied and unoccupied windows. Vael's default schedule runs 18:00 to 06:00 unless a custom schedule is set. The trade-off is that the schedule and the escalation path both have to be configured deliberately.

Comparison: back to base monitoring and the camera-side models

Model	Human dispatch?	Capital needed	False-alarm load	Best for
Back to base via graded centre	Yes (graded operators)	Low	Reaches operators, billed per event	Contract, insurer or police-response requirements
Outsourced remote video	Yes (live)	Low	Medium	High-risk sites needing live eyes
Automated verification only	No	Low	Filtered by AI	Lower-risk sites where notification is enough
AI verification (Vael) plus your escalation path	On escalation	Low	Largely filtered upstream (typically 90%+)	Unattended yards and depots with existing cameras
AI in front of a graded centre	Yes (graded operators)	Already owned	Largely filtered upstream (typically 90%+)	A1 and A2 stations carrying alarm volume
Self-monitoring with filtered alerts	No	Very low	Filtered by AI	Small sites and tight budgets
Scheduled AI and graded response	Scheduled	Low	Largely filtered upstream (typically 90%+)	Clear occupied and unoccupied windows

How to choose: decision criteria

1. Start with the contract and the policy. If a tender, lease or insurance schedule names graded monitoring, that settles it. Everything else is a design decision after that requirement is met.
2. Risk profile. High value or high consequence sites justify live human verification and a guaranteed attendance. Routine sites are often well served by verified automated notification.
3. Where the false alarms are landing. Every unnecessary signal that reaches an operator or triggers an attendance is billable. Filtering upstream protects both the customer's bill and your margin.
4. What "a response" means to the customer. Write it down and have them agree to it. Automated notification and certified dispatch are different promises, and customers should be buying the one they think they are.
5. Licensing. Match the model to your licence and your state's rules. Software filters and verifies; licensed parties dispatch.
6. Recurring revenue goals. Monitoring sold under your own brand turns one-off installs into monthly recurring revenue. See how CCTV installers build recurring revenue in 2026.

Adding a camera verification layer: what it involves

If verification on existing cameras is the piece you are missing, the setup is short. For the two delivery models side by side, see how installers offer CCTV monitoring. In short:

1. Point the cameras at Vael. It works with most IP cameras and NVRs via standard image upload (SFTP or FTP), including Hikvision, Dahua, Axis, Bosch, Hanwha Vision, Uniview, Honeywell, Avigilon and Pelco.
2. Configure detection rules, schedules and alert channels per camera.
3. Define the escalation path for verified events: keyholder, patrol, or a graded centre for dispatch.
4. Brand it and price it. See pricing, quoted per site.