Recording a failure
The driver (or a dispatcher) marks the job failed with a reason from your list — your industry preset seeds sensible reasons such as recipient unavailable, wrong address or access issue, and you can add your own. The customer can be notified with what happens next.
Automatic re-attempts
With automatic re-attempts on, a failed job is cloned into a fresh unassigned job scheduled for the next business day in your timezone, keeping the attempt chain (attempt 2 of 3). When the attempt cap is reached — three by default — no further attempt is created and your operators are alerted to arrange the return instead. Configure this in Dispatch → Settings → Re-attempts.
Presets whose customers cannot reschedule — medical, food and auto parts — never re-attempt automatically, whatever the toggle says: a failed run is marked for manual handling so a person decides what happens next.
Hold for pickup
If your operation holds failed deliveries for pickup instead, record that as your policy in the operation profile and turn automatic re-attempts off in Dispatch → Settings → Re-attempts. The engine does not read the policy answer yet, so with the toggle on it would still schedule next-day attempts.