What is PagerDuty?
PagerDuty is the dedicated incident-response platform for on-call rotations. It owns the escalation policy: who gets paged first, who gets escalated to after N minutes of no acknowledgement, what runbook fires, and how the incident is orchestrated from open to post-mortem. Engineering and SRE teams have used PagerDuty for years to absorb the operational risk of always-on systems, and ecommerce ops teams increasingly use it to do the same for revenue-critical paths. PagerDuty is not a chat tool. It is a contractual surface, where pages, schedules, and SLAs are first-class objects.Why Vortex IQ uses PagerDuty as an alert channel
Vortex IQ Nerve Centre detects revenue-at-risk events in real time: a checkout error spike, a payment processor degradation, a 5xx storm during peak hour, a stock-out on a top-10 SKU mid-campaign. These are not “post in a channel and hope” events. They need an on-call to acknowledge within minutes, with automated escalation if they do not. Routing severity-tiered incidents through PagerDuty gives Vortex IQ access to the customer’s existing escalation policies, schedules, and override windows. The on-call experience matches what the team already knows. No new pager. No new schedule. Just Vortex IQ-detected incidents flowing into the same PagerDuty inbox as the rest of their infrastructure.What gets delivered
Vortex IQ pushes the following events to PagerDuty:- Severity 1 revenue-at-risk incidents (checkout failure, payment outage, top-SKU OOS during a paid campaign).
- Severity 2 KPI breach incidents (refund rate spike, conversion drop, fulfilment SLA breach).
- Infrastructure incidents inferred from observability connectors (Datadog, New Relic, Sentry).
- Connector health incidents when a critical data source goes silent.
- Incident updates (acknowledge, escalate, resolve) synced from PagerDuty back into the Vortex IQ incident timeline.
How it integrates
- Auth model: Per-service Integration Key (routing key) generated in PagerDuty, stored encrypted in Vortex IQ. One key per service so escalation policies can differ by domain (commerce, infra, marketing).
- Formatting: Events API v2 payloads with
summary,severity,source,component,group,class, andcustom_details. Deduplication keys reuse the Vortex IQ alert fingerprint to coalesce flapping into a single page. - Routing: Severity tier maps to PagerDuty service, which inherits the customer’s existing escalation policy. Sev1 routes to the always-on rota, sev2 to the trading-hours rota, sev3 emits as a low-urgency notification only.
- Delivery semantics: At-least-once with retry on 429/5xx, exponential backoff, and a fail-open audit log if the API is unavailable so no event is silently dropped.
- Acknowledgement: PagerDuty incident state (acknowledged, escalated, resolved) syncs back via webhooks into Vortex IQ’s incident timeline.
Operational diagnostics
PagerDuty is the most contractually critical of Vortex IQ’s alert channels, and is monitored accordingly. Tracked KPIs:- Event submission success rate per service over 24h and 7d windows.
- Median submission latency from alert generation to PagerDuty acknowledgement.
- Mean time to acknowledge (MTTA) and mean time to resolve (MTTR) per severity.
- Escalation rate (incidents that escalated past first responder).
- Routing key health (revoked or rotated keys, unmapped services).
- Webhook back-sync lag (time between PagerDuty state change and Vortex IQ ingest).