Approvals & alerts
Sign-off that routes itself. Alerts that never sleep.
Route approvals by role, user or field value — with budget escalation and auto-approve so nothing stalls. Fire alerts the moment a record changes or on a schedule, with severity levels, escalation chains and cooldowns that keep the noise down.
Approval workflows
Trigger, approver, action.
Define who signs off and when. Approvals chain across roles, escalate by budget, and auto-approve after a deadline so work never piles up behind a missing decision.
1
Trigger
status_change · record_create · field_change
An approval kicks off when a record's status moves, a record is created, or a watched field changes.
2
Approver
role-based · user-based · field-value-based
Pick who decides: a role tier, a named user, or an approver resolved from a field on the record. Over $10K → CFO.
3
Action
approve · reject · delegate
Approvers act, or it auto-approves after N hours. Status flows pending → in_progress → approved / rejected / cancelled.
Alerts
Three severities. One escalation chain.
Alerts fire from a record change or on a cron schedule, sorted by how loud they should be — and escalate on a timer until someone owns them.
Critical
Something needs eyes now — a machine down, a threshold breached. Pages the escalation chain until acknowledged.
Warning
A drift worth watching — falling OEE, a queue building up. Surfaces without the urgency of a critical page.
Info
A heads-up for the record — a run completed, a milestone hit. Logged and visible, no escalation required.
Multi-step escalation chains
Unhandled alerts climb a chain of steps with configurable delays between them — from the on-shift lead to the supervisor to whoever needs to know next.
Cooldowns & instance lifecycle
A cooldown stops the same alert re-firing in a tight window. Each instance moves through active → acknowledged → resolved → escalated, so an owned alert stops paging.
FAQ
Questions about approvals & alerts.
Every approval step names an approver — a role (e.g. anyone with the Manager tier), a specific user, or one chosen by a field value on the record. Steps can chain, so a record flows from a line lead to a plant manager and onward. Higher-value records can escalate further: set a budget threshold and a record over, say, $10K routes to the CFO automatically.
Each approval can auto-approve after N hours so a request never stalls indefinitely. Until then it sits in_progress; approvers can approve, reject, or delegate the decision to someone else. The request status moves pending → in_progress → approved / rejected / cancelled, and every transition is recorded.
Alerts have three severities — critical, warning, info — and fire from a record change (event) or on a cron schedule. A cooldown stops the same alert from re-firing in a tight window, and each alert instance has a lifecycle (active → acknowledged → resolved → escalated) so an acknowledged alert won't keep paging. If it goes unhandled, multi-step escalation chains with timed delays move it up the line.
Put sign-off and alerts on autopilot.
Book a demo and we'll wire an approval chain and an escalation alert onto your own records.