Multi-Site Video Surveillance: How to Run Every Location From One System

Multi-site video surveillance means managing the cameras at every location you run (stores, clinics, schools, warehouses or offices) from one system, with one login and one set of permissions. The most practical approach for most organizations today records video locally at each site and uses the cloud to manage, search and view it from anywhere. That keeps bandwidth low, keeps cameras recording when a site loses internet, and lets you add locations without building a server room or a VPN at each one.

This guide covers how the main multi-site setups differ, what to check before you commit to one, and how to move a mixed collection of sites onto a single system without replacing cameras that still work.

Three ways organizations run video across multiple sites

Most multi-site organizations did not design their camera setup. They inherited it one location at a time. Whatever the history, it usually looks like one of three models.

The central server model was the standard answer for years, and plenty of organizations still run video management software that way. This guide focuses on the third model, because it removes the central server without uploading every camera's footage to the cloud around the clock.

Why a patchwork of camera systems costs more than it looks

The cost of separate systems rarely shows up as a line item. It shows up as time. A regional manager who needs footage from three stores logs into three apps, each with a different timeline, export format and search tool. An incident that crosses locations, such as a pallet that left one warehouse and never arrived at the next, means pulling clips from two systems and lining them up by hand.

It also shows up in IT work. Every recorder needs firmware updates, failed drives replaced and user accounts kept current. When an employee leaves, someone has to remember every system that person could log into. And when a recorder at a small site fails quietly, nobody finds out until the day the footage is needed.

One system fixes this only if it also handles the reasons the patchwork exists in the first place: different camera brands at different sites, limited bandwidth at smaller locations, and local staff who need access to their own site and nothing else.

How cloud-managed multi-site surveillance works

Each site gets a recording appliance on its local network. Cameras stream to that appliance, which stores the footage and runs detection. The appliance connects out to a cloud service that handles everything that has to work across sites: the dashboard, user accounts, search, alerts and remote viewing.

What stays at the site and what goes to the cloud

In a hybrid design, full video stays on site by default. What travels to the cloud is lightweight: thumbnails, detection events, search metadata, device health and account data. Video streams out only on demand, when someone opens a camera, plays back a clip, runs a search or shares an incident. Some platforms also offer continuous cloud backup for organizations that want an off-site copy.

This matters because the smallest locations usually have the weakest connections. A clinic or a store on a business broadband line would struggle to upload around-the-clock video from 30 cameras and still run its phones and point-of-sale system. Local recording with on-demand streaming keeps that upload light until someone actually needs to watch.

What happens when a site loses internet

Ask every vendor this directly. With local recording, cameras keep recording to the on-site appliance during an internet or cloud outage. What pauses is the cloud-dependent work: remote viewing, cloud search, alerts and push notifications. When the connection comes back, the site's recorded footage is still there.

What your network team will ask

Network security often decides whether a multi-site project moves forward. Expect three questions:

• Does it need inbound ports, port forwarding or a VPN at each site? The answer you want is no. The appliance should make outbound connections only, typically over HTTPS on port 443.
• Can the cameras sit on an isolated network? If cameras only need to reach the local appliance, they can live on a separate VLAN with no route to the internet.
• How are updates delivered? Look for signed, over-the-air updates that do not need someone on site or a maintenance window.

What to evaluate in a multi-site video surveillance system

A demo at one site tells you little about site number 30. These checks do.

• Camera compatibility at every site. Ask which brands and protocols are supported, then test the oldest camera model you own, not the newest. ONVIF and RTSP support covers most IP cameras, but codec support and how much of the AI runs on third-party cameras vary by vendor.
• Where footage lives and how long it is kept. Retention depends on storage at each site, camera count, resolution and frame rate. Ask how it is sized per site and how incident clips are preserved past normal retention.
• Behavior on a weak or lost connection. Ask what each site sends to the cloud in normal operation and what keeps working during an outage.
• Permissions by location. A store manager should see their store, and a regional director should see their region. Check that access can be scoped by site or camera group and that single sign-on works with your identity provider.
• Search across sites. Find out whether investigators can work from one place or still have to search location by location.
• Doors and alerts in the same system. If access control is on your roadmap, check whether door events are tied to the video of that door automatically, and which alerts come out of the box.
• Administrative load. Ask who patches the appliances, what happens when one fails, and how quickly a replacement comes back online with its settings intact.
• Evidence handling. Check export formats, shareable links for law enforcement or insurers, and whether viewing, exports and permission changes are written to an audit log.

With answers to these in hand, a side-by-side comparison gets much easier. Our roundup of the best cloud-based VMS platforms covers the main options.

How to roll out one system across mixed-vendor sites

Start with the sites that cause the most trouble, prove the system there, then standardize. A phased rollout usually runs in five steps:

1. Inventory each site. List camera brands, models, resolutions, recorder type, internet connection and who currently has access. This list drives server sizing and shows which cameras, if any, genuinely need replacing.
2. Pilot at two or three representative sites. Include your oldest cameras and your weakest connection, not just headquarters.
3. Set up locations, roles and camera groups before the wider rollout. Getting permissions right at site three is far easier than fixing them at site thirty.
4. Move the remaining sites on their own schedules. With a cloud-managed system, adding a location mostly means installing the appliance and connecting its cameras, so sites can move around lease renewals, recorder failures or budget cycles.
5. Retire the old recorders once retention overlaps. Keep the old system readable until its stored footage ages past your retention requirement.

Keeping existing cameras is what makes this pace possible. Daleville Community Schools, for example, retired its Lorex NVRs without replacing a camera.

Multi-site surveillance by industry

Retail and franchise operators usually care most about loss and consistency. Store managers need their own store, while loss prevention needs every store in the same view so a pattern at one location can be checked against the others.

Healthcare networks add privacy to the list, which makes scoped access essential. Soar Autism Center runs 30+ cameras at each of its clinics across Colorado and Arizona, and its master-level clinicians use them to observe sessions and coach staff remotely instead of traveling to every site. "Coram didn't lock us into their cameras. That was a huge win," said Daniel Spiegel, SVP of Growth and Development at Soar. More on hospital and healthcare security cameras.

K-12 districts often have one safety team covering several buildings, plus administrators who should see only their own campus. A single system lets the district respond to an incident at any school from the same app. See how this works for school security cameras.

Warehouses and logistics operators use multi-site video to follow goods between facilities, for example confirming when a truck left one site and when it arrived at the next. More on warehouse security cameras.

How Coram handles multi-site video

Coram is an AI physical security platform that brings video security, access control, guest management and emergency management together, built to work with the cameras organizations already have. For multi-site operators, it follows the local-recording model above:

• Each site gets a Coram Point, an AI NVR appliance that records footage on site and runs detection. Each Coram Point is assigned to a Location in the Coram app, and Locations help control which users can see which cameras.
• Footage stays on the Coram Point. Thumbnails, detections, metadata and alerts go to Coram's cloud, and video streams only when someone views, searches, archives or shares it. Continuous cloud backup is available as an option.
• Coram Points connect outbound only over HTTPS on port 443, with no inbound ports, port forwarding or VPN, and cameras can stay on an isolated VLAN. Updates are signed and delivered over the air without downtime, and a replacement Coram Point pulls its configuration from the cloud instead of being rebuilt by hand.
• Coram works with 150+ camera brands over ONVIF or RTSP, so existing IP cameras can stay in place.
• Every location is searched and reviewed in the same app, with natural-language search through Discover and Deep Investigation. Alerts come from a set catalog that includes Firearm Detection, person and vehicle of interest, and license plate of interest.
• Admins see every location. Regular, Limited and Live-only users see only the camera groups assigned to them, and SAML single sign-on works with Okta, Microsoft Entra ID and Google Workspace.
• With Coram Access Control, door events are paired automatically with video from that door. Existing readers, locks and cabling can be reused, while the controller panel is replaced.
• Exports are standard MP4 files, shared links can be set to expire, and viewing, exports and permission changes are written to an audit log.

Coram is SOC 2 Type II certified and HIPAA compliant, and security documentation is available in the Trust Hub. If you are planning to bring several sites onto one system, see how Coram works for multi-location enterprises or book a demo to walk through your site list.