WestPoint · Video software · March Networks
March Networks comes from video in shops and bank branches. What it sells is not better recording: it is running hundreds of identical sites from one place and crossing what you see with what the receipt, the cash machine or the transaction says. If that is your problem, there is little competition here. If it is not, you are paying for half of what you are buying.
The picture: a hundred and eighty shops, or three hundred branches. One recorder in each, nobody technical in any of them, and a loss prevention department that does not know which of the three hundred something is happening in.
With that map, the customer's question is not «how do I see one camera better». It is another one, and it has two halves. The first: how do I know the three hundred are recording, without sending someone out to look. Almost all the video that is missing in a retail incident had been missing for weeks and nobody had noticed.
The second: how do I get from «I have the video» to «I have the list». A loss prevention manager cannot watch three hundred shops; what they can do is open a report telling them which forty refunds this week were done with nobody in front of the counter, with the video of each one a click away.
That is what this firm does, and it is in both places: a console that manages the fleet of recorders —status, firmware, configuration, users and remote access to every site without opening every network— and a data layer, which the firm calls Searchlight, that crosses the video with the transaction and produces exception reports.
And what marks out its ground, said up front: this is not a platform for a large campus with a shift-based control room and nine subsystems from other manufacturers. There are better answers for that on this same list, and we say so.
It is designed from the ground up for many small, identical sites, and that shows in every architectural decision.
What you buy here is not alarms: it is lists. And what it costs is not the licence. It is three things, and none of them is video.
What you get, to be concrete about it: a weekly report saying «these operations fall outside the normal», with the video of each one next to it. It is not an alarm at three in the morning: it is a list on Monday morning for a person who has to decide. In loss prevention that is a change of trade, not one more feature.
The first thing it costs is THE DATA. Somebody has to hand over the transaction feed, and that somebody is not the security department: it is IT and very often the till system supplier. There is a list of supported systems; your version and your particular integration are a different question, and it gets answered with the till supplier in the meeting, before quoting. The most common reason one of these projects sits still for six months is the data. It is never the video.
The second comes before everything else and is not technical: crossing transactions with video is, in practice, watching how one particular person works. In Spain that touches data protection and it touches employment law — people have to be informed, you have to be able to justify that the measure is proportionate, and the workers' representatives have to be involved. We do not say it to make things difficult: we say it because that is the order in which this goes well, and because a system built without that conversation ends up being switched off where it hurts most.
The third is who reads the reports. An exception report nobody opens on Monday is a licence paid for nothing. This needs a loss prevention person with time set aside, not a name in the minutes. If that person does not exist yet, the honest recommendation is to buy the video layer and leave the data layer until they do — and we do recommend that, even though it is half a sale.
And a warning about expectations: exception reports flag what falls outside the norm, not what is guilty. Most of the lines will have an explanation. The value is in going from three hundred shops to forty lines, not in the forty being thefts.
The video layer is licensed like any other: by recorder and by channel, and it is predictable. The fleet console and the data layer are licensed by site, and sometimes by till or by operator position.
That has a consequence that changes the whole comparison and is worth saying out loud: here the bill grows with the number of SHOPS, not with the number of cameras. It is the opposite of every other platform on the list, and comparing quotations without looking at that is comparing two different things with the same name.
On top goes the annual maintenance, which here weighs more than on others because it is what keeps the fleet console and the firmware updates alive. Without it, the centralised management —which is the reason for the purchase— gradually falls behind.
And the line nobody budgets for, which is about hardware and about the calendar: three hundred recorders bought in the same year die in the same year. Renewing the fleet is a five-year plan in stages, and putting it in the table from the start is what avoids the year in which three hundred units have to be changed at once.
Of these six it is the one that fits or does not fit most clearly. There is almost never a middle ground.
You have a lot of similar sites and nobody technical in them: shop chains, branch networks, restaurants, petrol stations, pharmacies. If the video has to be crossed with the till, the cash machine or the transaction. If there is a loss prevention department with people in it. And if knowing which site has stopped recording is a real problem that nobody solves today.
You have one or two large sites with a control room: here the fleet console has nothing to manage and gets paid for all the same. If what you expect is powerful scene analytics, because that is not its ground. If a lot of subsystems from different manufacturers have to be integrated. Or if the till data is not going to arrive, because then half the product will never be switched on.
That there is somebody who takes ownership of the reports and somebody who hands over the transaction data. Both people, with time set aside. If they do not turn up during the sale, the project is delivered well and eighteen months later it is a very well managed fleet of recorders — which is something, but it is not what was paid for.
In a three-hundred-site project the value is not in installing one: it is in the three hundred coming out the same.
There are three parties involved here and not two, and the third is the one that decides whether the project lives: loss prevention.
This platform has analytics that are very useful and very much of its sector, and it is worth being exact about what they are.
What comes with it, in the recorder and in the cameras: people counting, queue time, dwell zones, heat maps, detection in zones. These are the retail measurements, and in a chain they serve two purposes at once: security and operations. The same system giving the shop manager a number is what stops the video being nothing but a security cost.
And what really sets it apart is not detecting: it is correlating. Crossing the operation with the image is a kind of analytics no other platform on this list has out of the box, and it solves a problem that scene analytics does not solve.
What you add: third-party analytics and number plates by integration, with the usual warning and with a shorter catalogue than the open platforms have. It gets checked before quoting.
The ceiling: these are rules, counts and correlation with a piece of data. They detect what can be counted and what can be squared with a receipt. What does not arrive is understanding the scene: that an argument in the queue is turning nasty, that somebody is setting something up, that what is going on over there is odd even though there is no transaction at all. For that you would have to describe the world in thresholds, and you cannot.
And since what follows is a recommendation with an interest behind it, it goes declared first: in the same house there is IRIS Neural, the NVMS from Infinity Neural, the other company in the group. It is our product, we make it, and anyone who recommends their own thing without saying so loses the right to be believed.
They live together, and here the split is clean because they do not compete: March governs the cameras, the recording, the fleet of sites and the crossing with the business data; IRIS brings the understanding of the scene, which is the question the exception report cannot ask. And if what you need is the receipt report, this already does that and nothing else is needed: we say that just as plainly.
If your problem is a lot of identical sites, there is more than one answer in the index and they do not all cost the same: the difference is whether you need to cross the video with a piece of business data or only to manage the fleet.
Get started
Those are the two questions that decide whether this platform fits you. Tell us how many sites there are and what sizes, which till or branch system you use, what you would like to be able to pull out in a report and whether there is somebody in loss prevention who is going to read it. With that we tell you what can be done, what has to be checked with your till supplier and what conversation has to be had first on the employment side. If yours is one large site, we will tell you to look at another one.
