PagerDuty review

PagerDuty is a vendor-run operations platform delivered as a service, not a library, CLI or self-hosted incident server. It sits between monitoring and alert-producing systems and the people responsible for responding. Teams use it to detect incidents, route notifications, coordinate on-call response and automate operational work.

The platform combines incident management, Signal Intelligence, automation and incident learning. PagerDuty also positions Paige as an SRE Agent for AI-assisted triage, detection and resolution. Its stated model blends AI agents with people during investigation and remediation, then uses incident data to improve future response. In a typical stack, PagerDuty is the response and orchestration layer rather than the source of application telemetry. The homepage advertises more than 750 integrations and APIs and MCP as ways to connect it to other tools. That breadth can reduce glue work, but teams should map which systems send signals, which actions PagerDuty can trigger and which response workflows require paid plan allowances.

PagerDuty advertises APIs and MCP for extending its 750+ integrations. The homepage does not name a specific API style, SDK, programming language, CLI or plugin system. Check the vendor’s developer documentation for the API protocols, supported SDK languages, authentication model, event formats and versioning that fit your stack before committing to an integration.

The site links to a Developer Platform, an integrations library, a Knowledge Base and PagerDuty University. Those are the starting points for finding implementation guides and supported connections. The integration catalogue is the clearest published signal of breadth: PagerDuty says it connects with hundreds of tools out of the box and lists more than 750 integrations. That does not establish how much setup a particular monitoring, chat or ticketing workflow needs. Confirm the relevant integration’s directionality, configuration requirements and plan eligibility. The homepage also promotes coordinating incident response from Slack or an IDE, but does not name the IDEs or provide SDK-language details.

Free costs $0 and includes five users, one on-call schedule, one escalation policy and 100 SMS or phone notifications per month. It is a useful entry point, but those notification and team limits matter when putting a real rotation on it.

Professional costs $25 per user monthly or $21 per user monthly when billed annually. Starter is listed at $233.33 per platform per month, calculated from a $2,800 annual fee; it includes five seats plus five bonus seats, a 30-seat cap, 5,000 Signal Intelligence Events and 1,000 AI Actions. Essential is $1,000 per platform monthly, based on $12,000 annually, with 25,000 events, 5,000 AI Actions and 2,000 Incident Workflows. Plus is $4,000 monthly, based on $48,000 annually, with 150,000 events and 16,000 AI Actions. Ultimate has custom pricing. Professional add-ons include AIOps, PagerDuty Advance, Runbook Automation, Status Pages and Live Call Routing. The pricing lists credit packs and additional Starter seats at 40 credits per seat; overage prices are not published.

PagerDuty is a commercial product, not an open-source on-call package. The product and pricing pages present paid plans and hosted platform features; they do not present a self-hosted deployment option or an open-source licence. Teams that require control of the incident service itself should ask PagerDuty about deployment and data handling before adopting it.

The homepage does not describe data export formats or migration tooling. Ask the vendor what incident history, schedules, escalation policies and workflow configuration can be exported, and whether those exports remain usable outside PagerDuty. Migration effort will depend on how much of the response process relies on PagerDuty-specific integrations, policies and automations. Reddit users discuss switching and pricing, but their comments do not establish how complete or difficult an export is. Treat portability as a procurement question: test the export path, identify replacement integrations and keep critical response procedures documented outside the service.

The plan allowances make production sizing a first-order task. Check expected users, Signal Intelligence Events, AI Actions, Incident Workflows and phone or SMS notifications against the plan before routing a full incident rotation. PagerDuty’s homepage does not publish API rate limits or detailed service-level terms; confirm limits, availability commitments, security controls and compliance documentation with the vendor for your deployment.

Reddit comments show differing expectations. One commenter values reliability, predictability and a stable interface during overnight incidents; others report switching after price increases or complain about scheduling and confusing workflows. These are individual opinions, not performance measurements, but they point to practical evaluation areas: test schedule changes, escalation paths, notification delivery and incident handoffs with the people who will use them. Ask how usage beyond included allowances is handled and what additional seats or add-ons cost. Do not assume the advertised event and action allocations cover peak incident activity.