You connect enterprise video with business systems by turning video into structured events and data, then moving that data through APIs, event-driven actions, and integration layers into the systems where work happens. Video becomes a source of real-time physical context: enterprise resource planning (ERP) and warehouse management systems (WMS) learn what is physically happening with goods, Internet of Things (IoT) platforms and supervisory control and data acquisition (SCADA) systems gain visual confirmation of machine and site states, and AI agents get the visibility they need to act. The connection works in both directions, so business systems can also send events and commands back to the video platform.
Why does video need to connect to business systems?
Business systems run on records. A WMS records where inventory should be. An ERP records orders and shipments. SCADA records sensor values and machine states. All of them describe physical operations, but none of them can see those operations. Cameras can. When video stays inside the security stack, that visibility is wasted. When it connects to business systems, it fills the gap between what the record says and what is actually happening.
What can video add to each type of system?
| System | What video adds | Example |
|---|---|---|
| WMS | Visual confirmation of inventory, picking, and dock activity | Confirm a pallet was staged at the right dock |
| ERP | Physical verification of orders, shipments, and receiving | Attach video evidence to a received shipment |
| SCADA / industrial control | Visual context for machine and process states | Confirm a valve or conveyor state when a sensor alarms |
| IoT platforms | Visual data alongside sensor telemetry | Pair a temperature alert with footage of an open cooler door |
| Building management | Occupancy and space usage | Adjust HVAC or lighting based on people counts |
| Ticketing / ITSM | Automatic ticket creation with video attached | Open a facilities ticket when a spill is detected |
| Incident management | Real-time events with location and evidence | Route a safety incident to the nearest supervisor |
| Digital twins | Live positions and states of people, vehicles, and objects | Update a site model with real-time activity |
| AI agents | Real-time physical context to reason and act on | An agent checks dock status before rescheduling a carrier |
What are the main integration patterns?
Most video-to-business-system integrations use one or more of four patterns.
1. Push events out
The video platform sends an event to an external system when something happens. In Nx EVOS, the rules engine can send HTTP GET or POST requests to third-party systems when an analytics event or other trigger occurs, with event details such as camera, event type, and object name. It supports Basic, Digest, and Bearer authentication, and custom headers can be added to each request.
2. Receive events in
An external system sends an event to the video platform, which then responds. Nx EVOS supports HTTP generic events, so a WMS, access control system, or IoT platform can trigger rules such as bookmarking video, moving a PTZ camera, or alerting an operator.
3. Query through APIs
An application or agent requests data or video on demand. The Nx Toolkit includes the Server HTTP REST API, which gives developers access to nearly every feature available in the Desktop Client: querying servers, cameras, and users, pulling live or recorded video, generating events, and creating rules. It supports GET, POST, PUT, PATCH, and DELETE requests and works with any programming language. The toolkit also includes Cloud and WebSocket APIs.
4. Build native integrations
Developers build plugins that run inside the platform. Using the Metadata SDK, a plugin can attach metadata to video and send custom events to the system. Developers can also embed custom web pages in the Desktop Client and interact with them through the In-Client Browser JavaScript API.
How do AI agents fit into video integrations?
AI agents are a new kind of integration endpoint. Instead of a fixed rule mapping an event to an action, an agent can interpret events, combine them with business data, and decide on a next step within its permissions.
A typical flow:
- The video platform detects a truck waiting at a dock longer than its scheduled slot.
- The event goes to an agent with the truck's location and wait time.
- The agent checks the WMS for staged orders and the dock schedule.
- It recommends reassigning a crew, or takes that action if it's within its permissions.
- The action and outcome are logged.
What should you plan for before integrating?
| Consideration | Why it matters |
|---|---|
| Data mapping | Video events need to map to business system objects such as orders, assets, or locations |
| Location model | Cameras and zones should map to the same locations business systems use |
| Event volume | Filter and aggregate events so business systems aren't flooded |
| Security | Use authenticated connections and scoped credentials |
| Latency | Match delivery speed to the use case |
| Ownership | Decide which team maintains each integration |
Where does an EVOS fit?
An Enterprise Video Operating System (EVOS) is the layer between cameras and the rest of the enterprise. Nx EVOS manages devices, interacts with AI, structures detections as metadata, and exposes them through APIs, SDKs, and a rules engine. That positions video as one more data source in the enterprise stack, rather than an isolated security system.
FAQ
Can video analytics integrate with a WMS or ERP? Yes. Video events can be pushed to these systems through HTTP requests or APIs, and the systems can query video data on demand.
What is a video analytics API? It's a programmatic interface that lets applications retrieve analytics events and metadata, or submit them to a video platform.
Can external systems trigger actions in the video platform? Yes. Nx EVOS supports HTTP generic events that let external systems trigger rules.
Do I need custom development to connect video to business systems? Simple integrations can use the rules engine's HTTP actions. Deeper integrations use the Nx Toolkit APIs and SDKs.