Support teams adding video escalation to a contact center inherit the same compliance question voice recording already answered: who decided to record, and who was told. Video raises it louder because the capture is richer — faces, screens, co-browse sessions — but the answer has the same shape: the decision is the tenant's, and a platform earns its place by making that decision expressible, never by deciding it for you. This post maps the two controls that carry the posture — which sessions record, and how participants are told — onto the video and co-browse controls Devotel Orbit ships, following the same tenant-owned framing the call-recording consent rules post uses for voice.
Why video recording isn't a platform-set policy
The recording decision sits above every jurisdiction's consent split (one-party in most of the US, all-party in a dozen states, lawful-basis plus transparency under the GDPR), and above the carrier and enterprise questionnaire layer that reads your recording posture as part of your standing. A platform that hard-codes a recording policy — records everything, or records nothing — just chose for you, and the choice it makes is rarely the one your counsel signs. On Orbit, every video capture control is tenant-owned and defaults to your configuration: eligibility, announcement, selection, retention. The platform ships the knobs; the answers are yours.
The two controls that carry the posture
Voice recording reduces to two orthogonal axes. Video and co-browse map onto the same pair:
Axis 1 — which sessions record (eligibility). Set recording off, on, or per-queue. On Orbit's video channel a session is recorded only when your configuration says so — a host can start, pause, and resume capture mid-session, and individual participants can be excluded from the capture. Pausing when a payment field or health detail comes on screen, and excluding a supervisor's muted channel from the file, are the configuration-level equivalents of the denied-consent filter the voice rules post describes.
Axis 2 — how participants are told (announcement). The visible indicator, the pre-join notice, whatever your policy requires. Orbit does not mandate an announcement mode; it exposes the control and the enforcement point where the join token is minted, so a joiner admitted while a recording is active sees the indicator your tenant chose to show. The audit trail per session — admission, mute, lock, removal — is what an enterprise questionnaire asks to see, and it is reconstructable from the tenant's session records.
Where co-browse adds a second layer
Co-browse (agent-side screen guidance during the session) is not a capture: the agent sees the customer's screen live, and no recording need exist. It becomes a recording question only when the session in which it ran is also being captured. The tenant-owned set therefore splits into two toggles:
- Co-browse eligibility — which agents may guide screens, and during which kinds of sessions. This is an agent-permission and per-queue question, same shape as the video eligibility axis.
- Capture of the co-browse session — if the room records while guidance is active, the guidance surface is part of the capture by default; if the tenant's policy excludes co-browse from capture, the exclusion control on the recording eligibility axis covers it.
The support team's posture should cover both. A team that records sessions but wants co-browse guidance to stay outside the capture configures the exclusion; a team that wants the full guided session on file configures inclusion. Orbit ships both paths; the answer is the tenant's policy, not a platform-default the program inherits blindly.
Retention and sharing as tenant-owned too
Regulated programs tend to fail compliance review on retention, not capture: the recording exists because a queue was configured to record; it is still accessible because nobody revoked the share link. The video recording controls carry the same tenant ownership through the lifecycle — pause-ability carries into which segments land in the file, and revocable share links carry out of it. Age the files against a retention setting your policy names (or a default the program deliberately accepted), and revoke links when the recipient's need ends. Both are per-recording tenant operations on Orbit, never a platform-side expiration you cannot override.
Where this posture lives on Orbit
The video recording and co-browse controls sit on the Video API inside the RTC-PaaS pillar, layered on the same moderation and token-mint enforcement the buyer checklist scores on the feature side. The recording attaches to the account's contact profile the session belongs to, so retention and access resolve in the same dashboard the other channels use.
Frequently asked questions
Does Orbit set a recording policy for us?
No. Every video and co-browse capture control is tenant-owned and is set by the tenant's configuration — eligibility per session or queue, the announcement mode shown to participants, the exclusion of individual participants, and the retention window on the archive. Orbit ships the controls; the compliance decision they encode is the tenant's, matching the posture the voice rules post describes.
Is co-browse a recording, or a live viewing?
Co-browse is a live agent-to-customer screen guidance capability — it helps the agent see and guide the customer's screen while the session runs. Whether the session in which it ran is also recorded is a separate eligibility question. A tenant's posture names both: which agents guide, and whether those guided sessions capture.
When do regulated programs usually fail compliance review on video?
On retention and sharing, not on capture. A queue-level recording decision is easy to document; what fails review is a stale share link or an archive with no age limit. The tenant-owned revoke and retention operations on the recording lifecycle are the controls that answer the questionnaire.
Published 21 September 2026.