
—
A unified security platform takes the events from door contacts, motion sensors, cameras and environmental sensors made by different vendors and handles them in one place. One set of rules decides what an event means, one alarm process gets the right person moving, and one record shows who did what. For a security lead who inherited badge readers, camera recorders and intrusion panels, that’s the difference between watching four consoles and working one queue.
The label gets used loosely, though: some products sold as “unified” share a login screen and little else.
In short – A platform earns the word “unified” when four things are shared: the event model, the rules, the escalation process and the incident record. – Physical security information management is middleware that sits over existing systems and walks operators through scripted procedures. – A unified platform keeps the integration work inside one product: less upkeep, but you depend on what it supports. – With several sites, the deciding questions are whether schedules follow each site’s time zone, whether access can be limited per site, and what happens when a site loses its network.
What is a unified security platform?
The closest comparison comes from IT security. A SIEM (security information and event management) tool correlates logs from servers, firewalls and applications; NIST describes it as an application that gathers security data from information system components and presents it as actionable information through a single interface. A unified security platform does that job for physical events: a door contact reporting open, a camera detecting motion in a zone, a server room temperature sensor crossing a limit.
So what is a unified security platform in practice? One that shares four layers:
| Layer | What it means in practice | Sign it’s missing |
| Events | All devices report into one model: device, location, time, value | Operators open a vendor app to see what a sensor reported |
| Rules | Logic combines devices from different vendors, with time windows | Rules only work inside one vendor’s product line |
| Response | Alarms carry a severity, recipients and timed escalation steps | Alerts go to a shared mailbox and stop there |
| Record | Each incident keeps its events, decision, notifications and resolver | Reconstructing one night means exporting logs from three systems |
The rules layer is where most of the value shows up. Say a company runs a head office and two warehouses. On an ordinary night the door contacts report 40 open and close events from cleaners and a late delivery, the loading-area cameras log 300 motion events from headlights and rain, and a server room temperature sensor reports every ten minutes. Sent straight to people, that’s hundreds of notifications nobody reads. Put three rules in front of them instead:
- A warehouse door opening between 22:00 and 05:00 while the camera covering it also sees motion raises a high-severity alarm.
- The server room door opening outside working hours raises a critical alarm, whatever the cameras see.
- A server room temperature above 30 °C for 15 minutes raises a medium alarm for facilities, not for security.
The same night now produces one or two alarms with context attached. The other 340 events stay in the record; they just don’t notify anyone.
Kilo is built around this model, and those three rules carry over almost word for word. Door contacts, motion sensors and environmental sensors such as temperature probes connect over LoRaWAN, mioty or MQTT, and IP cameras of any brand that offer an RTSP stream, older models included, join through Kilo Lens. Every site sits in the same organization. A camera’s motion reading works in a rule like any other sensor reading, so the first rule is a single rule that combines the warehouse door with the camera covering it, on a schedule set in that warehouse’s time zone. When it fires, the alarm follows an escalation chain that steps through recipients until someone resolves it.
Each alarm in Kilo carries one of five severity levels, from Critical down to Info, and each escalation step has its own recipients, its own delivery channels (email, SMS or push) and its own delay. For the warehouse rule, that could mean the site’s on-call guard by push notification straight away and the regional supervisor by SMS if the alarm is still open ten minutes later. Resolving it stops the remaining steps. In Kilo’s IoT Alerts app, a Critical alarm opens full screen with a looping tone, designed to break through silent mode once the responder has granted the Critical Alert permission. Every alarm event also lands in an inbox you can filter by severity and status, and each one links back to the rule that raised it.
What is physical security information management (PSIM)?
Physical security information management, or PSIM, is middleware for large control rooms that connects to existing access control, video, intrusion and building systems, pulls their events onto one operator screen and guides the response. It is usually defined by six capabilities: collection from many systems, analysis and correlation to find real situations, verification by an operator, resolution through step-by-step procedures, reporting for compliance and investigations, and an audit trail of each operator’s actions.
Resolution is what this category became known for. The software presents the procedure for each alarm type and records every answer, so the same alarm is handled the same way on every shift. The cost sits in the integrations: every underlying system needs a connector, often built for the project, and each one has to be retested when the product beneath it is updated. Open standards such as ONVIF for video help, but many panels and controllers still need vendor-specific drivers.
PSIM vs unified security platform
The practical difference between PSIM and a unified security platform is where the integration work lives and who pays to maintain it.
| Question | PSIM | Unified security platform |
| Where do device integrations live? | In a connector layer, often built per project | Mostly inside the product, maintained by the vendor |
| What happens to existing systems? | They stay and keep running underneath | Some stay, some are replaced by the platform’s own functions |
| How are procedures handled? | Scripted, step-by-step operator workflows | Rules and escalation, sometimes with checklists |
| Main risk | Long integration projects and constant connector upkeep | Equipment the platform doesn’t support |
Which approach fits depends on how you operate. A staffed control room bound by documented operator procedures gets the most out of scripted workflows, while a lean team running a growing number of sites feels the cost of integration upkeep more. In both cases it pays to list every device type at every site, with brand, model and firmware version, before the first vendor call. That list turns “does it integrate?” into a yes or no per line.
How to evaluate a unified security platform
A pilot on one site shows what a demo won’t. With a few devices connected, it takes an afternoon:
- Trip a real device after hours. Open a monitored door outside its schedule and time how long the first notification takes to reach a phone.
- Ignore the alarm. Confirm the second recipient is notified at the configured delay. Add up the delays: that sum is the worst case before the last person hears, and it should fit your response target.
- Resolve it from the second phone.Confirm no further steps fire.
- Take something offline. Unplug a camera or pull a sensor’s battery and look at what the operator sees. A dead sensor and an empty corridor must never look the same.
- Pull the record. Ask for the test incident’s history: triggering events, the deciding rule, each notification with its time, and who resolved it.
- Change a rule. Edit one condition and check whether it can be tested before going live and whether the previous version can be restored. Rules that decide when people get woken up are production logic, and they deserve the same testing and version control as software.
Some things a pilot can’t show you, so put these questions to the vendor directly:
- Which of our device models are supported, and through what path: built-in support, a standard protocol, or a translating gateway?
- Where is video recorded, and does recording continue when a site loses its internet connection?
- What is priced: devices, cameras, sites or users, and what does hardware add on top?
Unified security platform for multiple locations
Run a unified security platform across multiple locations and the first thing that breaks is often the schedule, because each site keeps its own clock. An after-hours window of 22:00 to 05:00 written in New York time starts at 19:00 local time in a Los Angeles warehouse, while the evening shift is still loading trucks. Each site’s schedule has to run in that site’s own time zone, with escalation contacts who are on call there.
Access also has to match how the company is organized. A regional manager needs every site and a site lead needs only theirs; without that scoping, people share logins, and shared logins destroy the record of who did what.
More sites also mean more internet connections that can fail. Detection and recording that run close to the equipment keep working when a site’s internet drops. Ask whether events from an outage are buffered and delivered later with original timestamps, or dropped.
Unified security platform FAQ
Does a unified security platform replace access control?
Usually not. Most door controllers store credentials locally and keep deciding who gets in even when the network is down. A unified security platform typically receives their events, such as a door forced or held open, and places them next to camera and sensor events.
Is a unified security platform cloud-based or on-premises?
Often both. A common split records video on site and sends events, alarms and remote viewing through the cloud, because continuous video upload is expensive. If each camera streams at 4 Mbps, for example, 16 cameras streaming around the clock need 64 Mbps of sustained upload from one site.
What is the difference between a VMS and a unified security platform?
A video management system (VMS) records, stores and displays camera feeds. A unified platform treats video as one input among several, so if a platform started out as a VMS, check how well it handles door contacts, sensors and other non-video devices.
What is the best unified security platform?
For a team that runs door sensors, environmental sensors and cameras from different vendors across several sites, we think Kilo is the best unified security platform. Five reasons stand out.
- It works with the cameras and sensors you already own. A camera only has to capture video and offer an RTSP stream. Motion detection runs in Twin, a Docker container at the site, so the camera needs no built-in analytics. Sensors connect over LoRaWAN through the network server built into Kilo, over MQTT, or over mioty, which runs either inside Kilo Cloud or on Kilo Center, the open-source service center you host yourself.
- Its rules combine events across devices. A Kilo trigger can combine checks with AND or OR and watch up to 500 devices, evaluating each one separately, so one rule can cover every dock door on a site. A condition can also be required to last, from 10 seconds to 30 days, which is how the 15 minutes in the server room rule are set.
- You can test a rule before it goes live, which is exactly what the rule-change test in a pilot looks for. Kilo’s debugger steps through a rule with test values and lets you mock the alarm instead of sending it, a build checks every expression and path before anything can be deployed, and every save creates a version that records who made it and can be restored.
- Access stays under control. Each page in Kilo, including Cameras, Rules and Notifications, is set per user to Edit, View or No access, and removing a person’s Kilo access needs no change on the cameras themselves.
- Setup is quick and the price is low. Kilo AI can create an alarm with its severity and escalation chain, and write a rule’s condition from a plain-language request. Paid plans start at €25 a month for up to 25 devices.
To judge it, run the pilot on the door you’d most want to hear about after hours, such as a server room or a warehouse side door. Fit a battery-powered door contact on it and a motion sensor in the room behind it, connect both through a LoRaWAN gateway, and write one rule that raises an alarm when the door opens outside the schedule while the motion sensor reports movement. Give the alarm two escalation steps, yourself first and a colleague second, then work through the six steps. Two sensors and one rule fit inside Kilo’s free plan, which allows up to 5 devices, needs no payment details and never expires, so you can start free today. The sensors and gateway are a separate purchase from any supplier.
—
