Matt Fishback, Global Technology Partner Manager & Americas Team Lead, Milestone Systems explains how to make the move from hardware to cloud in retail security.
Retail has always been a balancing act between safety, service and sales. Today, the stakes are higher than ever.
Organized retail crime is on the rise, labor shortages have left fewer staff to cover the floor and customers expect shorter lines and smoother experiences.
Shrinkage alone accounted for more than $112 billion in losses in 2022, according to the National Retail Federation, with much of it tied to organized theft.
Faced with these pressures, retailers are turning to technology in new ways. The story of modern retail video is no longer just about catching shoplifters on camera.
It’s about transforming retail security and store operations through cloud connectivity, AI-driven analytics and shared intelligence across locations.
If Amazon Go’s “just walk out” model is a glimpse of what lies ahead, the real story is in the steady migration from heavy local hardware to flexible, open platforms that make AI practical today.
For years, video systems in retail security looked the same. Each store had its own set of cameras connected to a local server tucked in a back room.
Security staff used the recordings mainly for investigating theft after the fact. That model worked in its day, but it was costly, fragmented and hard to scale.
Large chains with dozens or even hundreds of locations had little choice but to duplicate this set-up store by store.
Updates were inconsistent, hardware required constant upkeep and valuable insights from one location rarely made it to another.
The result was a patchwork of isolated systems, all churning out video data but offering little beyond reactive review.
That’s changing. Today, more retailers are centralizing operations through cloud and hybrid systems that connect multiple locations into a single, manageable environment.
A hybrid approach to retail security is especially valuable in areas with limited bandwidth, where some data stays on-site while the broader system remains connected to the cloud.
This shift cuts down on hardware costs, makes scaling far easier and allows corporate IT teams to set consistent standards across the enterprise.
At the center of this retail security transition is open platform video management software (VMS).
Unlike proprietary systems that lock users into a single brand, open platforms allow the integration of cameras, sensors and analytics from many different vendors.
That flexibility is crucial for an industry as dynamic as retail, where new threats and technologies constantly emerge.
Open systems give retailers the confidence that they can add new hardware, software and cloud capabilities quickly without scrapping what they already have.
The first priority for any store is safety and security. Organized retail crime has grown into a professional, and often violent, enterprise.
Thieves target high-value goods in bulk, from electronics to designer clothes and move them quickly into online marketplaces.
Traditional cameras can only show what happened after the fact. AI-driven video analytics, tied into an open video platform, give retailers new ways to stay ahead.
Some stores are experimenting with device tracking using WiFi and Bluetooth signals.
If a phone or other device was present during a prior theft, the system can flag its return and alert staff to keep a closer watch.
Others are using smart cabinets that rely on anonymous facial recognition to grant or deny access to high-value merchandise.
GPS-enabled packaging and forensic spray markers add more layers of deterrence and evidence collection.
Video analytics can also spot behaviors like loitering near exits, groups gathering or someone carrying tools that don’t belong in a store environment.
All of this is made more effective by an open platform VMS. Instead of piecemeal systems that don’t talk to each other, retailers can connect these diverse technologies, from cameras to Internet of Things (IoT) sensors, into one ecosystem.
When an alert triggers, it can be cross-checked with video, access data and point-of-sale information. In this way, retailers turn scattered data points into actionable intelligence.
But retail security is only part of the story. Video analytics are also transforming how stores operate on a day-to-day basis.
Heat maps reveal which aisles attract the most traffic. Dwell time analysis can show where shoppers linger, helping managers decide where to place high-margin products.
AI can spot empty shelves, misplaced items, spills or long checkout lines before customers start complaining.
In short, retail security video data is becoming a tool for customer experience as much as for crime prevention.
Staffing challenges make this even more important. With fewer associates on the floor, managers need a constant flow of information to know where help is needed most.
Video platforms can generate alerts when carts are abandoned, aisles are blocked or checkout lines exceed a set threshold.
By identifying anomalies in real-time, AI acts like an extra set of eyes, allowing store staff to focus on service rather than monitoring screens.
Shared infrastructure pushes the concept even further. At a large Las Vegas mall, for example, the plan was to offer a subscription package of analytics that individual retailers could tap into as needed.
Airports are exploring similar models, using shared systems not only to enhance safety but also to study customer movement through shops and restaurants, insights that can guide tenant placement and lease pricing.
For smaller retailers who might never afford enterprise-grade analytics on their own, these shared platforms provide access to tools that were once out of reach.
The pattern is clear: AI-driven video is moving retail security beyond theft prevention to become a core business tool.
And with open platforms at the center, retailers can keep adding new layers as both threats and opportunities evolve.
If today’s migration to cloud and AI is the foundation, the Amazon Go model shows one possible destination.
These small-format stores, often in airports and stadiums, are packed with cameras and sensors. Shoppers grab what they want and simply walk out, with their accounts billed automatically.
Behind the scenes, the system tracks hand movements, item selection and even whether a product is put back on the shelf.
The experience feels almost magical: no lines, no waiting, no checkout. But the reality is complex.
Amazon has poured years of research and development into this system, and the camera density alone makes it expensive to scale widely.
The technology continues to expand in airports, stadiums and corporate campuses, showing that frictionless shopping has a place in the retail ecosystem.
For mainstream retailers, the lesson is not that every store will look like Amazon Go tomorrow. The takeaway is that customers value frictionless experiences and that the industry is heading steadily toward greater automation.
Cloud and open platforms will be the bridge that makes it possible for retail security.
By connecting cameras, sensors and analytics in a flexible environment, retailers can trial new ideas, learn from the data and adopt what works without having to start from scratch each time.
It helps to see this change as a migration path. The first stage was the era of heavy, hardware-bound systems in every store.
The second is the cloud and hybrid approach that’s gaining ground today.
The third will bring increasingly autonomous stores, where AI takes on the monitoring and anomaly detection while people focus on strategy and service.
The exact form of retail security operations may differ from one store to another.
But, the overall direction is unmistakable: smarter, faster and more connected retail environments built on AI and open video management software.
This article was originally published in the October edition of Security Journal Americas. To read your FREE digital edition, click here.
