Transfers, forwarding and delegation flows

The transfer family

• Blind (basic) transfer: A hands the call to C and leaves immediately. New signalling dialog + fresh media path B<->C; A's leg dies. Failure mode: C never answers and the call is simply lost — which is why front-desk staff should use…
• Safe transfer: like blind, but if C fails to answer the call RETURNS to A. The failure mode becomes a feature.
• Consultative transfer: A talks to C first (B on hold hears music), then completes. Two concurrent calls at A until completion — brief double-media load at A is expected, not a leak.
• Transfer to phone number: B gets re-targeted to the PSTN — on Direct Routing this spawns a NEW outbound leg through B-side routing policy; a user who 'can't transfer to mobiles' has a routing/policy gap, not a transfer bug.

Forwarding and simultaneous ring

Forwarding re-targets BEFORE answer (immediate, or unanswered-after-N-seconds); simultaneous ring clones the alert to a second target with first-answer-wins. The chain rules that keep support sane:

• Chains evaluate at each hop's settings — A forwards to B, B forwards to C works, but loops are detected and killed by the service (a 'call never arrives' with a forwarding loop in settings is the classic).
• Sim-ring to PSTN consumes an outbound leg (cost + CLI implications: the callee's mobile sees the configured CLI treatment, confusing everyone the first week).
• Voicemail interplay: whoever's UNANSWERED timer fires first wins the call — sim-ring to a mobile whose carrier voicemail answers in 15s beats a 20s Teams timer every time. That mystery is carrier voicemail racing Teams, and it is ALWAYS diagnosed backwards from 'the call went to my personal voicemail'.

Delegation (boss-admin)

Delegates answer/place calls AS the boss: shared line appearance, on-behalf-of CLI, held-call pickup across the pair. Configured per user (client or admin via calling settings), gated by AllowDelegation in the calling policy — the surface our MC1429012 verdict watches. Failure modes: delegation configured but policy off (silently non-functional) and the change-management one our mapping now alerts on: Microsoft shipping behavioural changes to the delegate experience.

What to watch (proofs)

• Which ballet happened: the user's Call Analytics timeline shows each leg — transfer/forward events appear as call segments with targets; read the segments before theorising.
• Timer races: Get-CsUserCallingSettings -Identity <user> — forwarding/ unanswered config incl. delegates and timers; compare with the mobile carrier's voicemail delay to predict every race's winner.
• Policy gates: Get-CsTeamsCallingPolicy | fl AllowCallForwardingToPhone, AllowCallForwardingToUser, AllowDelegation per instance (scanned nightly; the delegation mapping already fires on drift).
• The lost-transfer audit: blind transfers that ended nowhere show as short C-legs with no answer in analytics — a weekly count of them is the case for retraining reception on safe transfer.