The idea is simple, have a primary respondent, and a secondary respondent.
Primary
- First point of contact
- We expect immediate response
- We expect full ownership of incidents to be taken
Secondary
- Backup, in case primary contact is unable to respond
- If the primary doesn’t acknowledge an issue within specified timeframe, secondary kicks in
- If the load on primary is too much, they step in
Misc
- Make sure to clearly define escalation policies