Interview Scheduling Coordinator Handoff: Change Owners Without Ghost Invites

Transfer one live interview workflow in eight steps

Use one finalist, one agency coordinator leaving coverage, one successor, an external client approver, four panelists, two scheduled sessions, one tentative hold, Google and Microsoft calendars, one unconnected interviewer, and a late candidate reply delivered after the ownership change.

1. 1

   Resolve the canonical candidate-stage record

   Start with search ID, candidate ID, application and stage, interview IDs, current revision, recruiter relationship owner, agency coordination owner, client approver, required and optional roles, approved substitutes, organizer calendars, provider event IDs, proposals, holds, participant messages, ATS references, and terminal proof. Do not reconstruct ownership from the newest inbox thread.
   • Every event, hold, and message maps to the intended candidate and stage.
   • Recruiter judgment and client relationship ownership remain explicit.
   • A changed ATS assignee is evidence, not proof that calendar and message ownership moved.
2. 2

   Separate the six ownership surfaces

   Record who owns recruiting judgment, routine coordination, client approvals, each participant relationship, each provider event, and every outbound sender. Add credential scope, delegate authority, effective time, expiry, escalation path, and disclosure class. One human may fill several roles, but the roles cannot collapse into one editable owner string.
   • The successor cannot inherit calendar or messaging authority by implication.
   • The existing organizer identity remains attached to each provider event until verified otherwise.
   • Private candidate context stays outside broad calendar and handoff notes.
3. 3

   Freeze commands from the old coordination revision

   Move the workflow into handoff-pending and stop stale workers from reminding, proposing, approving, booking, rescheduling, canceling, or adding attendees. Preserve late candidate replies, panel responses, and provider notifications as audited input. Reject side-effect commands whose expected owner or meeting revision no longer matches.
   • No duplicate reminder leaves from old and new coordinators.
   • A queued add-attendee command cannot alter the panel after the freeze.
   • The freeze pauses side effects without discarding valid participant evidence.
4. 4

   Inventory current calendar and conversational state

   Read exact Google and Microsoft events, organizer identities, attendees, times, zones, status, conferencing, and provider permissions. Join those facts with connected free/busy and bounded replies from participants who never connected a calendar. Mark each item current, stale, inaccessible, moved, canceled, duplicated, or unknown.
   • Calendar connection remains optional and helpful.
   • Missing access and silence remain unknown rather than free.
   • A late reply is matched to participant, meeting revision, source, zone, freshness, and consent.
5. 5

   Authorize the successor and define the provider plan

   Verify successor identity, employment or contract status, client-approved contact path, sender address, calendar permissions, ATS access, retention scope, and escalation authority. For each existing event, choose an explicit path: retain the current organizer with valid delegate or service access, ask the authorized organizer to act, or replace the event through a controlled cancel-and-rebook operation when policy and provider behavior require it.
   • Ownership change never becomes a reason to bypass provider permissions.
   • Cancel-and-rebook requires current participant evidence and approved disclosure.
   • A provider-specific limitation returns an owner and next action instead of a guessed workaround.
6. 6

   Rebind future outreach without restarting the panel

   Create a new coordination revision with the approved sender, reply routing, reminder caps, local-time windows, and escalation owners. Keep current candidate and panel windows until they change or expire. Tell participants about a contact change only when necessary, using the minimum context needed to respond or correct the record.
   • The candidate is not asked to resubmit valid availability for internal coverage convenience.
   • The client receives only the handoff facts its role requires.
   • Repeated replies and worker deliveries resolve idempotently under the new revision.
7. 7

   Repair active proposals, holds, and invitations

   Invalidate approvals tied to unavailable organizer access or stale participants. Refresh required Google and Microsoft free/busy, keep current conversational windows, recompute overlap in named IANA zones, and route substitutions or deadline changes to the right human. Issue one versioned reschedule, cancellation, or booking command and reconcile uncertain provider responses before retrying.
   • Required attendance wins before optional preference.
   • A timeout cannot create a second interview or erase the wrong event.
   • The recruiter keeps candidate and client judgment while the workflow handles settled coordination rules.
8. 8

   Verify control and return one handoff receipt

   Read back every in-scope event and verify organizer, time, zone, candidate, required panel, optional decisions, conferencing, and invitation state. Confirm old workers and reminders are closed, new replies route correctly, ATS references point to the current coordination revision, and every unresolved permission or organizer issue has one named owner.
   • The successor can identify the current state and safe next action without inbox archaeology.
   • No active hold, reminder, proposal, or invitation is orphaned.
   • Completion is verified handoff or one bounded exception, not an updated assignee field.

Compare coordinator handoffs by what remains controlled

The winning workflow changes internal ownership without making candidates repeat work or leaving panel invitations attached to abandoned credentials and queues.