What is video retention and who sets it?
Retention is the length of time recorded video remains available. Most systems record in a loop: once storage is full, the oldest footage is overwritten by the newest. The retention period is therefore the amount of time that fits into the available storage, or a fixed limit set by policy, whichever is shorter. A short period saves storage and limits the privacy footprint, while a long one gives more time to discover and investigate a past incident.
The right period depends on how quickly you are likely to notice a problem. A site watched in real time may only need footage for a short review window, while an unattended property that is checked occasionally needs longer, because a theft may not be discovered for days. Privacy rules, contracts, and insurer requirements may also set minimum or maximum periods. Because these obligations differ by place and industry, confirm with counsel or the local authority.
How do bitrate, cameras, and recording mode determine storage?
Storage needed is roughly the video bitrate multiplied by time, multiplied by the number of cameras. Bitrate is the amount of data per second the camera produces, and it rises with resolution, frame rate, and scene complexity. A busy scene with moving leaves, rain, or noisy low-light footage compresses poorly and needs more storage than a calm daylight scene. Modern compression such as H.265 reduces bitrate for the same visual quality compared with H.264.
Recording mode changes the equation as well. Continuous recording stores everything and gives a complete timeline, but uses the most space. Event-based recording stores only when motion or an analytic trigger occurs, which stretches retention considerably, though it risks missing the lead-up unless a pre-event buffer is kept. Many systems store a high-quality stream during events and a lower-quality one otherwise. Choose the mode by asking what you most need to prove later and what you can afford to lose.
Edge storage versus cloud storage: where should footage live?
Edge storage means the recording happens on a drive in or near the camera or on-site recorder. Its great strength is independence: footage is captured even when the network is down, and no bandwidth is consumed. Its weakness is that the footage is only as safe as the hardware. If the equipment is stolen, damaged, or fails, the recording goes with it, and retrieving video requires a connection or a visit.
Cloud storage keeps a copy off site, so footage survives damage to the unit and can be reviewed from anywhere. Its limits are bandwidth and cost: uploading continuous high-quality video from many cameras over a cellular link is demanding, so systems often upload events, clips, or lower-quality streams. The common answer is both. Record locally for completeness, upload selected footage to the cloud for safety, and understand exactly which footage ends up where and for how long.
How should video be exported so it works as evidence?
Evidence has to be believable. That means exporting footage as close to the original as possible, rather than recording a screen or re-encoding it to a smaller file. Preserve the original timestamps and camera names, include the relevant time before and after the event, and note whether the clock was accurate. Many systems offer a native format with an included player and a standard format such as a common video file, and it is wise to keep both.
Integrity matters too. A cryptographic hash, which is a short fingerprint calculated from the file, lets anyone later confirm that the file has not changed. Watermarks and export logs add further assurance. Keep a record of who exported the clip, when, and for what reason, and store the original export unaltered while working from copies. Requirements for evidence differ by jurisdiction and agency, so ask the receiving party or counsel what format they want before you send anything.
What are chain of custody and legal hold?
Chain of custody is the documented trail showing who had the evidence from the moment it was collected until it is presented. Every handoff is noted: who exported it, how it was stored, who received it, and when. Gaps in that trail can allow someone to question whether the footage was altered. Systems that log exports automatically and restrict access to authorized users make this documentation much easier to maintain.
A legal hold is an instruction to preserve footage that would otherwise be overwritten. When an incident occurs or a dispute is expected, the relevant footage should be protected right away, before the loop erases it. Good systems let you lock a clip so it is exempt from deletion. Also consider the opposite obligation: deleting footage on schedule can be required by policy or privacy rules. Confirm the correct approach with counsel or the local authority.
How should you evaluate a retention and export setup?
Ask for the real retention period under your actual settings, not a theoretical maximum. Ask how it changes if you add cameras or raise resolution. Find out how footage is locked against overwrite, who can do so, and how quickly. Test an export before you need one: pick a past moment, export it, play it on a different computer, and check that timestamps and quality are intact. A system that cannot do this smoothly will be a problem during a real incident.
Ask about the copies. Where is the second copy stored, who can reach it, and how is it protected? What happens to video if you cancel service? Can you retrieve footage in a standard format? Review who has permission to export, and make sure exports are logged. Finally, put the policy in writing: how long footage is kept, who may request it, and how requests from outside parties are handled.
Key terms
- Retention period
- The length of time recorded video remains available before it is overwritten or deleted, set by storage capacity or policy.
- Bitrate
- The amount of video data a camera produces per second, which rises with resolution, frame rate, and scene complexity.
- Pre-event buffer
- A short rolling recording that lets event-based systems save the seconds before a trigger as well as after it.
- Hash
- A short digital fingerprint calculated from a file; if the file changes in any way, the fingerprint no longer matches.
- Chain of custody
- A documented record of everyone who handled a piece of evidence, from collection through presentation, with times and purposes.
- Legal hold
- An instruction to preserve specific footage from deletion or overwriting because it may be needed for an investigation or dispute.
- Native export
- A copy of footage in the recorder's original format, preserving quality and metadata, often with a bundled player.
A trailer like this keeps a local copy of video on the unit, so recording continues even if the connection drops. On VDS trailers, edge recording and 2 weeks of cloud recording are standard value-adds, and the data plan has unlimited data with no overages, so uploads do not create surprise charges. Verified incidents handled by the operations center can be exported with supporting video.
Common questions
How long should I keep security video?
It depends on how quickly you would notice a problem, plus any contract, insurer, or privacy requirements. Unattended sites usually need longer than watched ones. Because obligations vary, confirm with counsel or the local authority, and document the policy you choose.
Can I just screen-record footage as evidence?
That is not recommended. A screen recording loses original quality, timestamps, and metadata, and is easier to challenge. Use the system's export function, keep the original unaltered, and ask the receiving agency which format it prefers.
What is the difference between edge and cloud recording?
Edge recording saves video on hardware at the site, so it works without a connection but depends on that hardware surviving. Cloud recording keeps a copy off site, which protects against theft or damage but depends on bandwidth. Using both is common.