WestPoint · Video software · Qognify · Hexagon
This firm does two things: video management and PSIM. A PSIM does not record images: it sits above every system and directs what the operator has to do, step by step, leaving it written down. It plays on different ground from the other five, and it is the easiest one to buy badly.
The word sounds like an acronym and the idea is simple: the program that tells the operator what to do, and that notes down what they did.
Picture a control room with eight different systems: video, doors, intruder detection, fire, perimeter, radio, lifts and plant. Something goes off at three in the morning: three screens have to be checked, you have to decide whether it is real, make a call, open a door and then write it up.
A PSIM turns that into a sequence. It takes in the alerts from every system, brings them together, opens the right camera and lays out the steps in order: check this, ring this number, confirm that. And it logs every step with its time and with who did it.
The problem is not technical, it is about people: the right procedure lives in the head of the experienced operator, and when it is the new one on a Sunday, it does not.
This firm has both pieces: a video management line —which comes from several products with histories of their own coming together, Ocularis and Cayuga among them— and a PSIM line. Since 2023 it has belonged to Hexagon, which also makes emergency and mapping software.
Two layers. The video one looks like the others. The PSIM one looks like nothing else, and that is where nearly all the work comes in.
The video layer is what you would expect: a management server, recording servers and clients. It scales by adding recording servers, and its decisions are those of any VMS: cameras per server, storage, what gets duplicated and what retention.
The PSIM layer records nothing: it connects to each system with its own connector and above them it runs procedures. That means it does not have a size in cameras: it has a size in connected systems and in written procedures.
And here is the most useful truth on this page: in a PSIM project the product is 20 % and the connectors and the procedures are 80 %. Writing what to do for each alert is weeks of work with the head of security and with the shift operators, and if nobody takes ownership of that the PSIM ends up being used as a very expensive alarm viewer.
The second warning is the classic problem with every PSIM: connectors have versions. IT patches the access control system, the connector does not support that version, and the control room stops seeing the doors. It cannot be avoided, it gets managed with a list of supported versions.
Integrating is the product, not a feature. That is why the conversation is a different one.
It connects with access control, intruder detection, fire, perimeter, public address, radio, plant systems and with video, its own or somebody else's. When a customer has nine systems from six manufacturers, the other platforms on this list do not compete on that ground.
The question that tells apart someone who has done a PSIM from someone who is about to try: for each system, which connector is there, what version has it been tested with, what can it do —only read alerts or also send commands— and who maintains it.
A connector that reads alerts and cannot open a door is half of what the customer thought they were buying. And a bespoke one has to be maintained every time either side changes: fifteen years of installation are fifteen years of maintenance.
The video layer, like any VMS: by camera channel and by tier. Predictable and easy to budget for.
The PSIM layer is not. It is licensed by what gets integrated and by how many people use it at once: connected systems and concurrent positions. The day a tenth system is added —a fence, a building, the lifts— that is a licence line, not an afternoon.
It is the year-three surprise and it is avoidable: all it takes is drawing up the list of systems for the next three years at the start. On top goes the maintenance, which here weighs more because it is what gives you updated connectors: without it, the PSIM gradually goes deaf.
And the cost that is not licence and is the biggest one: the hours of writing the procedures. A budget where that line is small ends in a renegotiation or in a PSIM used as a viewer. We put it in with its figure and we defend it.
Of the six it has the sharpest border. Inside its own ground it has little competition; outside it, it is a cost that is hard to justify.
A real control room, with shift operators, a lot of systems from different manufacturers and procedures that have to be followed the same way at four in the morning: airports, ports, utilities, transport, large campuses. And anyone who has to prove to somebody else what was done and when.
A single site with forty cameras, one guard and no written procedures. The cost is in the hours of writing procedures that do not exist and in maintaining connectors for three systems that were already visible from the VMS. For that, the video layer on its own is the answer.
That there is somebody on the customer's side who takes ownership of the procedures. Not a name in the minutes: a person with time set aside. If they do not turn up during the sale, the project is delivered well and degrades within eighteen months, and then we recommend not buying a PSIM yet.
In the video layer, what you do with any VMS. In the PSIM layer, work that looks more like organising a control room than installing software.
It is the platform of the six where the split matters most, because there is a third thing that is neither IT's nor security's: the procedures.
IT keeps the infrastructure of both layers, plus a responsibility that has to be assigned to them explicitly: never update any connected system without checking the versions matrix. That is what stops the control room losing sight of the doors on a Monday.
Security keeps the operation of the video layer —users, permissions, retention, exports— and that of the control room: which alerts arrive, with what priority and at which position.
The procedures belong to neither of the two, and that is the gap where these projects die. One from two years ago quoting a phone number that no longer exists is a trap: the operator follows it, it fails, and they lose confidence in the whole system.
You can carry quite a lot on your own: creating simple procedures, adjusting priorities, changing phone numbers. Somebody has to be trained, because if every number goes through the integrator the catalogue goes out of date.
What is left to the integrator is the connectors, the version jumps coordinated across both layers, the procedures with logic between systems, and diagnosis when something does not arrive. With nine manufacturers inside, that is half the value of the contract.
Training: the longest of the six, with three audiences. The operators, by shift and with real simulations. The owner of the procedures, several days, because they are going to edit them. And IT, on the versions matrix and why it exists.
Nobody buys this platform for its analytics, and understanding that is what avoids a disappointment.
In the video layer there is what you would expect: motion detection, zones, basic rules and the ability to pick up the alerts the cameras generate on their own. Useful, and the same as almost any VMS brings.
Where this firm puts its work is not in detecting, it is in what happens afterwards: any alert enters a procedure, opens the right camera, lays out the steps and leaves a record.
You can add third-party analytics, number plates, counting, perimeter, and there is a real advantage: because the PSIM is built to connect things, new analytics is one more connector and not a rebuild.
The real limit shows up here in a particular shape: it is excellent at directing what has to be done when something is detected, and it depends entirely on something having detected it. A PSIM with poor detection is a very well organised control room that never switches on.
That is why this is the platform where what comes next makes the most sense, and we declare it first: in the same house there is IRIS Neural, the NVMS from Infinity Neural, the other company in the group. It is our own product, and if we did not say so everything above would lose its value.
They live together: the VMS and the PSIM govern the cameras, the recording and what has to be done; IRIS brings the understanding of the scene, which is what generates the alert that did not exist before. And if the detection you have covers what worries you, what you are short of is procedures.
If what you need is to record well and see well, the other platforms in the index are the conversation. A PSIM is for when the problem is the control room, not the camera.
Get started
If the answer is two, you do not need a PSIM and we will tell you so. If it is seven from five manufacturers, with shifts and with procedures that today live in people's heads, it is worth talking. Tell us what systems there are, how many positions the control room has and whether there is somebody who can take ownership of the procedures.
