Connecting channels

Connect a customer channel to the right team, validate both directions, and monitor health.

Expected outcome: Messages arrive in the intended inbox, approved replies leave through the expected sender, and failures are diagnosable.

1. Connect a channel as an operating workflow
A provider connection is complete only when routing, ownership, delivery evidence, and recovery are proven.

2. Confirm provider ownership and administrative access
Use a client-owned business identity. Keep credentials and verification codes out of screenshots, tickets, and shared notes.
Review point: Never borrow a personal sender for production work

3. Choose the team inbox before connecting the channel
Define membership, default routing, assignment expectations, and the destination for unmatched conversations.
Review point: Every inbound message needs a safe default route

4. Complete the provider connection in Settings
Follow the provider flow, return to WRKZY, and read the resulting connection state instead of assuming success.
Review point: Connected, degraded, and broken require different action

5. Send an authorized test message into WRKZY
Verify sender identity, contact matching, team inbox, queue, timestamp, and owner assignment.
Review point: Test with a clearly marked demo contact

6. Reply through the same customer thread
Confirm the correct sender, recipient, session or template rule, and provider-backed delivery event.
Review point: A submitted request may still fail later

7. Review health and delivery evidence
Settings health, conversation history, and delivery state should identify the failed layer without exposing credentials.
Review point: Record time, state, and safe identifiers for support

8. The channel has a team, an owner, and a recovery path
Document who monitors connection health and who handles delivery exceptions before real volume begins.

This silent visual walkthrough was screen-recorded from the sanitized Acme Retail demo workspace. Follow the linked written guide for exact prerequisites, role boundaries, edge cases, and recovery.