An alarm nobody can see
is an alarm nobody is coming to.

The problem verification solves

An intrusion alarm makes a decision with almost no information. A detector saw movement, so the panel sent a signal, so the monitoring centre called somebody. Nobody in that chain knows whether a person is in the building or a balloon drifted past a PIR. The same question decided differently on a job site is alarm or cameras for a construction site.

That is why false alarm rates are what they are, and it is why police departments have spent two decades adding conditions to alarm response. The end of that road arrived in Seattle on 1 October 2024. SPD now responds to alarm-company calls only when there is evidence of a crime in progress, and a sensor or motion activation on its own does not qualify1.

Read that again if you own a building in Seattle. Your alarm going off is not, by itself, a reason for anyone to come.

What changes with verification

The operator stops guessing. The alarm arrives with evidence, photos or a clip, so a person looks at the actual event before deciding what to do with it.

That cuts both ways, and the second way is worth more than people expect. A verified alarm gets reported as a person in the building. An unverified one that turns out to be a tarp in the wind gets cancelled at the desk instead of becoming a false dispatch, a municipal fine, and one more reason for a department to deprioritise your address.

Photo verification: what our Ajax alarms send

We put MotionCam detectors where a picture settles the question: the stock room, the office, the yard. A door contact on its own sends no picture. A MotionCam is a motion detector with a small camera in it. When it trips, it sends a short series of photos with the alarm, the first within about nine seconds2. The operator at Becklar, a UL-listed central station, sees what set it off. So do you, in the Ajax app.

Photos are not video. They show a person or an empty room, which is usually enough to tell a break-in from a balloon, but they do not show the seconds before the trigger. Seattle's dispatch rule names audio, video, a panic alarm and an eyewitness as evidence, and photos are not on that list. Where that matters, the next tier adds video.

Video verification: how CHEKT does it

A CHEKT bridge sits between the alarm panel and the cameras. It is wired to the panel's zone outputs, and each zone is mapped to the camera that covers it.

When a zone alarms, the bridge pulls video from that camera and passes it to the monitoring platform. It buffers the last seconds of video continuously, so the clip starts before the trigger rather than after it, which is usually the difference between seeing someone arrive and seeing an empty frame where they used to be.

Cameras and recorders need to speak ONVIF or provide a stream the bridge can reach on the local network3. Most modern IP cameras and NVRs do. The bridge does not replace your alarm panel, it does not replace your recorder, and it does not turn an analogue camera into an IP one.

What the operator sees

With photos: the account, the zone that alarmed and the photo series from the detector. With video, the list below.

The account and the site. The zone that alarmed. The camera mapped to that zone. A clip including the buffered pre-alarm seconds, and live video where the installation supports it.

What they do with it follows the central station's procedure and your local authority's rules. The honest limit is that a clip showing a person forcing a door is strong verification, and a clip showing an empty yard is not. Verification is only as good as whether a camera was pointed at the thing that alarmed.

Where your city stands

Washington does not have one rule, and this is the part most vendors skate over.

Seattle: since 1 October 2024, supporting evidence of an in-progress crime is required for dispatch. Sensor-only alarms do not qualify1.

Pierce County: deputies are dispatched on two separate alarm trips plus two attempted calls to the user or an alternate contact, and successful contact is not required4.

Kent, Tacoma, Auburn, Federal Way, Renton and the rest: we could not confirm current dispatch standards from official sources, so we are not stating them. These rules are set city by city and they change. Ask us and we will check your city with you before you buy anything.

Which cameras actually need video

Not all of them, and we will say so before you ask.

Verification belongs on restricted areas: the yard after hours, the storage compound, the plant room, the roof, the controlled area, the gate nobody should be at on a Sunday. Those places have a simple rule behind them, which is that nobody has a reason to be there. That makes the operator's job fast and the alarm worth acting on.

A camera on a public entrance is the opposite. It sees legitimate people all day, so verifying it produces a stream of events that mean nothing, costs you a lens fee each month, and trains everyone in the chain to stop caring. The rest of your cameras keep recording exactly as they always did; they are just not part of the verification loop.

In practice most sites verify a handful of lenses out of a much larger camera count. That is the design question we settle at the site walk, and it is usually the difference between a monthly bill that makes sense and one that does not.

What it costs

Photo verification is part of the monthly alarm monitoring, with no per-camera fee. Video verification through CHEKT is about $125 to $150 per lens per month as of September 2026, quoted per site. The lens count is what moves it, which is the whole reason the section above matters: verify the restricted areas and the bill stays proportionate to the risk.

For context, the wider market runs from tens of dollars per camera for basic clip delivery up past $100 for services where a human being genuinely reviews the event. This sits in the second group.

What it is not

It is not a guard service. Nobody is watching your cameras continuously; the photos or the video arrive when a detector says something happened. For a live guard behind the cameras, see remote video monitoring.

It is not a guarantee of police priority. Some jurisdictions classify video-confirmed alarms as higher priority, and some will not dispatch without that evidence at all, but the decision belongs to the agency. Any vendor promising you a police response time is promising something they do not control.

And it is not a substitute for cameras pointed at the right things. A detector with no camera covering it cannot be verified, which is a design question we settle at the site walk rather than a licence you buy.

Where we work

From 415 Railroad Ave S, Kent, WA 98032: Kent, Renton, Tukwila, Federal Way, Auburn, Covington, Maple Valley, Tacoma, Puyallup, Sumner, Bonney Lake, Tehaleh, University Place and Gig Harbor. The Olympic Peninsula and sites farther afield are by appointment and quote.