Comparison
Monitored versus unmonitored alarm systems
The usual version of this page argues that monitoring is obviously better. The data from San Diego complicates that, and it is more useful to you than the sales pitch: the question is not really monitored against unmonitored, it is verified against unverified.
What each one is
Two different products, sold under one word
An unmonitored system has one mechanism: it makes noise on site and hopes somebody acts. On a commercial site — an industrial park at two in the morning, a warehouse on a Sunday, an office floor over a holiday weekend — that assumption usually fails. The building is empty, the neighbours are empty, and the siren runs its cycle and stops. Nothing is recorded and nobody is told.
A monitored system replaces hope with a procedure. When a device trips, the panel transmits a signal identifying the site and the zone that activated. A staffed central station receives it and works the response procedure agreed for your account: attempt verification, contact the people on your call list in the order you set, and request law enforcement dispatch where the activation stands up.
Self-monitored sits between them: the panel sends an alert to a phone, not to a station. It is a real option for sites with staff present through the hours that matter, and a weak one for the hours commercial sites are most exposed on.
What actually happens when an unverified alarm is dispatched
This is the number that should shape the decision, and it does not point where a security company's marketing usually points.
97.97%
DERIVEDComputed from SDPD's own published disposition codes for call type 459A. It is NOT an official SDPD false-alarm statistic, and a disposition of no crime outcome is not identical to the administrative false-alarm determination SDPD uses for permit and billing purposes — it includes cancelled calls and calls where no report was required. These are counts of CALLS DISPATCHED, not of alarm activations: activations resolved by a monitoring centre never reach SDPD and are not in this data.
| 2025 | Calls |
|---|---|
| Burglary alarm calls dispatched | 17,075 |
| Closed with no crime outcome | 16,728 |
| Report or arrest | 90 (0.53%) |
San Diego Police Department, Police Calls for Service, City of San Diego Open Data Portal. Period: calendar year 2025. Type: COUNT of dispatched calls; the percentage is derived. Retrieved: 2 September 2026.
That proportion has been remarkably stable, sitting between 97.2% and 98.1% every year since 2019. The full series, the disposition breakdown and the method are on false alarm statistics for San Diego.
The comparison
Monitored or unmonitored, decided on what the dispatch data shows
Monitor it if an activation has to produce a response; leave it unmonitored only if the system exists to record, not to summon anyone. The deciding fact is not the monthly fee. It is that an unverified activation is a weak instrument: in San Diego in 2025, 97.97% of dispatched burglary-alarm calls closed with no crime report and no arrest. Monitoring earns its cost when a human verifies the activation before anyone is asked to respond — which is what changes the quality of that call. An unmonitored system is a legitimate choice for a site that wants evidence after the fact and is not expecting a response at the time.
97.97% of dispatched burglary-alarm calls closed with no crime outcome
San Diego Police Department was dispatched to 17,075 burglary-alarm calls in the city in 2025 — roughly forty-seven a day. 16,728 of them closed with no crime report and no arrest; 90, just over half a percent, produced a report or an arrest. Derived from San Diego Police Department, Police Calls for Service, City of San Diego Open Data Portal, call type 459A, calendar year 2025. Retrieved 2 September 2026. This is our own calculation from published disposition codes, NOT an official SDPD false-alarm statistic — a disposition of no crime outcome includes cancelled calls and calls where no report was required, and is not the administrative false-alarm determination SDPD uses for permits and billing.
The proportion is stable, not a one-year artefact
It has sat between 97.2% and 98.1% every year since 2019, so the 2025 figure is the normal state of affairs, not an unusual year. Full series, disposition breakdown and method on our San Diego false alarm statistics page.
These are dispatched calls, not activations
Activations resolved by a monitoring centre never reach SDPD and are therefore absent from this data. The figure describes what happens once police are already on their way, which is precisely the situation monitoring is meant to prevent.
Monitoring is worth paying for when it verifies
The value is not that somebody watches a screen; it is that an activation is checked before a response is requested. That is what separates a call worth dispatching from the 97.97%. How verification works in practice: remote video monitoring.
Unmonitored is a real answer for a recording-only site
If the system exists so that footage can be pulled after an incident, and nobody is expecting anyone to arrive during one, monitoring adds a monthly cost against a benefit the site is not asking for. What footage is for after the fact: liability and incident documentation.
The choice is per site, not per company
A yard holding plant overnight and an office suite with a locked server room are different decisions, and a multi-site operator will often want both arrangements at once.
What we are licensed to monitor
Alarm monitoring is covered by BSIS ACO #7501. We hold no C-10 or C-16, so fire alarm monitoring is not ours and is coordinated with your own licensed fire contractor.
What it means
The argument this data actually supports
It would be easy to read that figure as an argument against monitoring, and easier still for us to ignore it. Neither is right. What it is evidence about is unverified response: an alarm signal, on its own, is a poor predictor of a crime in progress, and a system whose entire output is "a sensor tripped" is asking somebody to drive to your building on very thin information.
The United States Department of Justice reached the same conclusion from a different direction: its problem-oriented guide on false burglar alarms lists emergency police response to unverified burglar alarm calls among the responses it does not recommend.
So the useful axis is not monitored against unmonitored. It is whether anybody looks before deciding. Verification means a human being reviews live or recorded video from the sensor that triggered before anyone escalates, so a swinging banner, a delivery driver or a spider on a lens is resolved by looking rather than by dispatching a patrol car. That turns a binary signal into an actual assessment, and it is the thing worth paying for.
What we will not do is put a number on the improvement. There is no primary source establishing a reduction figure for verified response, and the figures circulated in this industry do not trace back to one. What can be said with a source is narrower and stated above.
Choosing
When each option is the right one
Unmonitored is the right answer when
- The site is occupied whenever it is at risk.
- The deterrent value of an audible siren is the actual objective.
- There is no asset whose loss would matter enough to justify a service relationship.
- The building already has monitored video coverage and the intrusion panel is a secondary layer.
Some sites genuinely do not need monitoring, and a contractor who cannot say that is not worth listening to on the question.
Monitored is the right answer when
- The building is empty for meaningful stretches.
- There is stock, equipment or data whose loss is material.
- An insurer or a regulator requires it.
- You need the timestamped record of activations and open/close events.
- Nobody on your staff can be relied on to act at three in the morning.
And in almost every monitored case
Add verification. Given what the dispatch data shows about unverified activations, a monitored system without a way to look at the zone that triggered is buying the weaker half of the product. Whether verification suits a given site depends on camera positions and sightlines, which is settled at survey, not promised in advance.
The service side of this is on commercial alarm monitoring, the video half on remote video monitoring, and the cost mechanics of an ongoing monitoring relationship on what drives commercial security system cost.
Scope
What we are licensed to monitor, and what we are not
Alarm and monitored CCTV work is BSIS-regulated and we hold the Alarm Company Operator licence, ACO #7501, which is the California licence covering the sale, installation, servicing and monitoring of alarm systems. The cabling, camera and intercom side sits under CSLB#976737, our C-7 low voltage contractor's licence. Both are verifiable with the issuing agency.
Our monitoring scope is intrusion, video and access only. We do not hold a C-10 or C-16, so we do not work on fire alarm systems; that is coordinated with your own licensed fire alarm contractor rather than performed or monitored by us. If a proposal you are comparing bundles fire monitoring with intrusion monitoring, that is a different licence and a different contractor. We work on commercial premises only and not on private residences.
One administrative point that attaches to the monitored option and not the unmonitored one: in the city of San Diego an alarm permit is required from the police department before a system is monitored, on SDPD Form 1067, and it is renewed annually. Sites outside the city fall under their own jurisdiction's rules.
| Monitored | Unmonitored | |
|---|---|---|
| What happens on an activation | An operator verifies before a response is requested | A local siren and a notification; nobody verifies |
| Who decides to call police | The monitoring centre, after verification | You, from a phone alert, usually without seeing the site |
| Evidence after the fact | Recorded, plus the operator log of what was seen and done | Recorded |
| Ongoing cost | A monthly monitoring fee | None beyond maintenance |
| Best fit | Sites where an activation must produce a response | Sites recording for evidence, not expecting attendance |
No pricing appears in this table. Monitoring is quoted per site because it depends on camera count, hours covered and the agreed response plan.
Common questions
Monitoring questions
Because the alternative is not "no false alarms", it is "nobody knows". An unmonitored system makes noise on site and stops. Nobody is told, no record is created, and if the building is empty at the time the activation is simply lost. Monitoring changes what happens after an activation instead of preventing the activation, and the second output matters as much as the first: a timestamped record of every activation, restore, open and close on your system. That record is what an insurer asks for after a loss and what tells you your night crew is disarming at the wrong hour.
RelatedCommercial alarm monitoring
It can be, and it is worth saying so plainly, not dismissing it. If a site has staff present through the hours that matter, a phone alert to a named person who will actually act on it is a real response path. Where it stops working is the case commercial sites are most exposed to: nights, weekends and holidays, when the alert arrives to somebody asleep or off-duty with no procedure behind it. Self-monitoring depends on a person; central-station monitoring depends on a procedure. Decide which your operation can rely on at three in the morning.
RelatedCommercial alarm monitoring
In the city of San Diego, an alarm permit from the police department is required before an alarm system is monitored, using SDPD Form 1067, and it is renewed annually. That is a live administrative obligation attached to the monitored option and not to the unmonitored one, and it comes with consequences for repeat activations. Other jurisdictions set their own rules, so if your site is outside the city of San Diego, check that city instead of assuming this applies.
RelatedCommercial alarm monitoring
No, and anyone telling you otherwise is overselling it. Verification changes what happens after an activation, not whether the sensor trips. A swinging banner, a delivery driver or a spider on a lens still triggers the detector; what changes is that a human being looks at video from the sensor that triggered before anyone decides to escalate. We are not going to attach a percentage to that improvement, because there is no primary source establishing a reduction figure for verified response and the numbers circulated in this industry do not trace back to one.
RelatedCommercial alarm monitoring