In projects moving an analog or HD-analog camera system to IP, the problem usually comes not from the cameras but from recording gaps: the old DVR is removed too early, a zone has no footage for a few days, and you find out at the first incident review. Replacing cameras is the easy part. The real decision is how much of the existing cabling, recorders and power distribution will be reused, which areas can’t tolerate downtime and over how many periods the budget will be spread. Below, we explain how to choose between hybrid, phased and full replacement, and how recording continuity is preserved throughout the migration, in the order we work on site.
Inventory first: what’s really on site
A strategy can’t be built without knowing what’s there. A common mistake is treating the DVR’s channel list as the inventory; the channel list doesn’t tell you where the cable runs, how many splices it has or what the camera is powered by. Gather the following for each camera point in a single table:
- Camera technology: classic CVBS (including 960H), HD-TVI, AHD or HD-CVI. The recorder’s channel info screen or the camera label often shows this. Two or three technologies running mixed on the same site is common; in systems that different companies have added to over the years, it’s almost the rule.
- Cable type and route: RG59/RG6 coaxial, coaxial combined with power cable (siamese) or UTP used with baluns. Approximate length, number of splices and connectors. This information isn’t written down without going into the ceiling space and looking.
- Power model: a central 12 V DC / 24 V AC power unit, or a local adapter next to the camera? Points with local adapters are the ones left in the dark during a power cut despite the UPS.
- Recorder: DVR/XVR model, channel count, hybrid (IP channel) support, disk capacity and actual retention achieved. The number of days visible on the timeline, not what the label says.
- Image adequacy: does the camera serve its purpose today? If an entrance camera doesn’t produce identification-level footage, that point is redesigned even if its cable is sound.
Classify the areas at the same stage. At points such as checkouts, the main entrance and shipping docks, recording interruptions aren’t acceptable; for a general overview camera inside a warehouse, a few hours of planned downtime are tolerated. The phase plan is built on this classification. A plan without classification stays on paper.
Migration options: hybrid, phased, full replacement
Migration with a hybrid recorder
With a hybrid DVR (XVR with some manufacturers) that accepts analog and IP channels together, new points are added as IP while the existing cameras keep running. The investment is split into small steps. There’s a price too: hybrid devices have limited IP channel counts, total incoming bandwidth and analytics support. A picture we often see on site is a device that has hit its IP channel limit, running half analog and half IP; from that point on, a second migration from the hybrid device to a VMS is unavoidable. A hybrid device is a stopover.
Phased IP migration
The IP infrastructure (PoE switches, camera VLAN, recording server or NVR) is built from the start according to the target architecture; cameras are replaced zone by zone. The old DVR runs in parallel until its own cameras are replaced. The budget is spread over periods without straying from the target architecture; in return, you have to monitor two systems together during the migration. With this option, write ONVIF Profile S and T support into the specification for camera purchases; even if the recording platform changes later, the cameras can be reused. Profile definitions are on the ONVIF profiles page.
Full replacement
Cabling, network, recording and cameras are renewed in a single project. It gives the cleanest result if the analog infrastructure is old, undocumented or faulty, or if there’s a window of opportunity such as a building renovation. The ceiling is opened once. In return, it needs teams working simultaneously on site and an outage plan written minute by minute.
Comparison and decision order
| Criterion | Hybrid | Phased IP | Full replacement |
|---|---|---|---|
| Initial investment | Low | Medium, spread over periods | High |
| Downtime risk | Low | Controlled, zone by zone | Depends on planning, highest |
| Cable reuse | Full at analog points | Selective (EoC or tested UTP) | Usually none |
| Fit with the target architecture | Weak, an interim solution | Strong | Strong |
| Analytics and integration | Limited by the device | Full, at IP points | Full |
| When it fits | Short-term budget constraints | Medium to large facilities running continuously | Old infrastructure, a renovation opportunity |
Build the decision order with three questions:
- Are existing cables undocumented, damaged or incompatible with target camera positions? If yes, full replacement, or at least a phased migration in which the cabling is renewed.
- Does the facility run 24/7 and can’t accept long outages? If yes, a phased migration; critical areas are moved with parallel recording and the greatest care.
- Is there budget for only a few new points, and does the existing DVR have a reasonable remaining life? If yes, a hybrid device can be a temporary solution; but put the target architecture in writing now, otherwise the temporary solution becomes permanent.
Which cables can be reused
Reusing cable is the biggest saving. The wrong decision also produces the most expensive fault: the camera drops now and then, nobody can find out why, and the team goes back to the same point for months.
- UTP used with baluns: If the cable really is Cat5e or above, in one piece and its length is within Ethernet’s 100 m channel limit, it can be used for IP cameras. The decision is made on the basis of a cable tester’s measurement; a continuity test isn’t enough. A line carrying analog video may not carry 100 Mbps Ethernet. Cables with CCA (copper-clad aluminum) conductors, which heat up and cause voltage drop under PoE, are ruled out.
- Coaxial: Doesn’t carry IP directly; it can be used with Ethernet over Coax (EoC) adapters. Distance, bandwidth and PoE limits need separate assessment; details are in the IP cameras over coax and EoC limits article.
- Power cables: Central 12 V DC lines usually become redundant after moving to PoE. If they won’t be removed, they’re labeled and de-energized; don’t leave live but ownerless cables in the ceiling.
- Conduits and trays: Existing routes can often be used when pulling new UTP or fiber; fill ratio and distance to power lines are checked.
For new runs, cable category selection follows the general rules of network infrastructure; the choosing Cat6, Cat6A and fiber guide covers this decision in detail.
A phase plan that preserves recording continuity
The risk most often skipped in a migration is the recording gap that forms while the old system is being removed, and losing access to old recordings. The order we apply is:
- Infrastructure phase: PoE switches, the camera VLAN, time synchronization (NTP), UPS and the recording platform are installed and tested. Not a single analog camera is removed at this stage.
- Pilot zone: A few points in a non-critical zone are moved to IP. Image quality, night performance, retention and the viewing client are verified together with users. Don’t check night images in daylight.
- Zone-by-zone migration: In each zone, the new IP camera is installed and put into recording first, then the analog camera at the same point is taken out of service. At critical points, having two cameras run together for a short time is an acceptable cost.
- Preserving old recordings: The DVR isn’t turned off after the last analog camera is removed. It runs isolated from the network and open only to authorized access until the recording retention period expires; when the period expires, the disks are destroyed according to policy. For setting the retention period, see the camera recordings and KVKK retention period article.
- Closeout: Central power units, baluns and unused cables are either removed or labeled and documented.
Budget items and commonly forgotten ones
Camera and recorder prices are the visible part of a quote. The items that push back the schedule are usually the following, and most aren’t in the first quote:
- Switches with an adequate PoE power budget and their UPS capacity (see PoE power budget calculation).
- Recording storage: IP cameras produce higher resolution and bitrate than analog; disk capacity is recalculated for the same retention period. The old DVR’s disk size isn’t used as a reference.
- VMS or NVR channel licenses, client licenses and annual maintenance fees.
- Cabinets, patch panels, labeling and cable test reports.
- Engineering time on the network side for VLANs, access rules and cybersecurity hardening.
- Keeping the old DVR running for the retention period and securely destroying its disks at the end.
- Privacy notices and warning signs to be updated if the coverage area changes.
- Operator training and documentation of the new system.
Post-migration verification
The migration of each zone is closed with measurable acceptance criteria. “There’s video” isn’t an acceptance criterion:
- Verified on the recording platform that each camera records with its planned stream profile (resolution, frame rate, codec).
- Actual retention compared with the target retention period.
- Night image checked on site, under real lighting conditions.
- All devices get their time from NTP; no drift between camera and recording server.
- Default passwords changed, unused services disabled, firmware versions recorded.
- Camera inventory updated with IP, MAC, switch port, location and recording server information.
- Playback and export tried together with the operator.
To clarify the scope, cable reuse decision and phase plan specifically for your site, you can see how we work on the analog CCTV to IP camera migration and survey and consulting pages.
Frequently asked questions
Can analog cameras and IP cameras be used on the same recorder?
Yes, hybrid DVR/XVR devices accept analog and IP channels together. However, the number of IP channels and total incoming bandwidth are limited; see a hybrid device as a transition-period solution, not a permanent architecture.
Do existing coaxial cables become completely useless for IP cameras?
No. With Ethernet over Coax adapters, IP cameras can be carried over existing coaxial lines. If the cable’s condition, distance, bandwidth or PoE needs aren’t suitable, running new UTP or fiber gives more reliable results.
What happens to old recordings during the migration?
The old DVR keeps running after the last analog camera is removed, isolated from the network and open only to authorized access, until the retention period expires. When the period expires, the disks are securely destroyed according to the retention and destruction policy.
Can UTP cables used with baluns be used directly for IP cameras?
They can if the cable really is Cat5e or above, in one piece and within the 100 m channel limit. The decision is made on the basis of a cable tester’s measurement; cables with CCA conductors are ruled out for PoE.