Sign in →
Operations1 min read

Operations — Overview

Internal observability and outbound integration. Alerts, ERP sync, payment reconciliation, metering health, the event log. Five pages, one job: keep the platform honest.

Updated 2026-06-30Suggest edits
Docs Operations Overview

Operations is the internal-facing reliability section. Alerts & Notifications pushes events out to your systems; ERP Sync and Payment Reconciliation keep the books straight; Metering Health and Event Log answer "is anything broken right now?". This is the section your on-call opens first when a customer reports something looks off.

What Operations is for#

Two clusters. Outbound integration — Alerts & Notifications (signed webhooks to your endpoints), ERP Sync (invoices pushed to QuickBooks / Xero / NetSuite), Payment Reconciliation (gateway → ledger match). Internal observability — Metering Health (per-billable-unit aggregate signals) and Event Log (every platform event, searchable + streamable).

INFO
Public reliability — Service Status, Incident Manager, Scheduled Maintenance, API Status — is in , not here. Operations is for what only you see; Storefront publishes what your customers see.

Pages in this section#

A simple incident playbook#

When a customer reports "billing looks wrong" or "events aren't arriving," walk the section in this order:

1
Event Log — search the customer + a recent timestamp
If events are landing at all, you'll see them here. Severity filter shows ERROR / CRITICAL in red.
2
Metering Health — is the relevant billable unit DOWN, DEGRADED, or NO_DATA?
If NO_DATA, the unit is declared but no events arrived in the window — usually an SDK / gateway misconfiguration upstream.
3
Alerts & Notifications — did our outbound webhook deliver?
Delivery Log shows FAILED + the response code. Retry policy reschedules; manual retry is one click.
4
ERP Sync — did the invoice make it to the accounting system?
Sync log surfaces per-invoice failures with the ERP-side error.
5
Payment Reconciliation — does the money match?
Open Payment Reconciliation last — it confirms what landed in your gateway matches what we billed.

Where to send alerts#

Alerts & Notifications ships 14 SYSTEM event types out of the box. Wire each one to the destination that fits your workflow:

Event familyTypical destinationWhy
INVOICE_GENERATED, PAYMENT_OVERDUEAR / collections inboxFinance acts on these — keep them in the system finance already lives in.
WALLET_LOW_BALANCE, USAGE_THRESHOLD_REACHEDCustomer success Slack channelPre-emptive renewal + top-up conversations.
SUBSCRIPTION_ACTIVATED, SUBSCRIPTION_CANCELLEDCRM webhookKeep your pipeline in sync with billing reality.
SERVICE_DEGRADATION, CREDENTIAL_DRIFT_DETECTEDPagerDuty + on-call channelReal reliability signal — page the right person.
WEEKLY_DROP_DETECTED, PACING_ALERT_PROJECTEDCS dashboard / weekly digestTrend signal, not page-worthy. Batch + review.
  • Storefront → Service Status / Incident Manager — the customer-facing side of reliability.
  • Intelligence → Ingestion — the read view on the same metering pipeline that Metering Health monitors.
  • Protocols → Webhook Ingestion — the inbound mirror of Alerts & Notifications (events FROM third parties).
  • Governance → Margin Guard — policy that causes alerts; this section is where the alerts land.