Moving from PagerDuty to Oppex: what changes for on-call and cost
Your paging tool does its job: it tells someone something is wrong. But most of the time in an incident is spent after the page — sorting real alerts from noise, finding the right person, opening dashboards and digging for the cause.
That is the part Oppex is built for. This guide covers what changes for your on-call engineers when you move from PagerDuty to Oppex, what it costs, and how to switch without betting production on it.
Where the time actually goes
Teams typically see over a thousand alerts a week, and roughly one in thirty needs a human. Someone finds that one by scrolling. Then the clock keeps running: getting the right people in and finding the cause usually take up most of an incident, while status updates for leadership are written by hand mid-incident.
Here is the same 2 a.m. incident — a bad deploy to checkout that fires 38 alerts in four minutes — handled two ways:
| With a paging tool | With Oppex |
|---|---|
| 02:14 Alert storm: 38 separate notifications | 02:14 38 alerts grouped into one incident |
| 02:20 Wrong person paged, re-routed by hand | 02:15 Right on-call engineer paged automatically |
| 02:35 Five dashboards opened to look for clues | 02:16 Nova: bad deploy v4.12, high confidence, evidence attached |
| 02:50 An executive asks for an update | 02:17 Rollback approved with one click in Slack |
| 03:05 War-room call with six people | 02:19 Leadership updated automatically |
| 04:24 Bad deploy finally found and rolled back | 02:23 Resolved; root-cause report drafted, owner assigned |
| Day 2 Postmortem pieced together from memory | |
| 6 people woken · 7 manual steps | 1 person woken · 1 approval |
The timings are an illustrative example, but the shape is familiar to anyone who has carried a pager: two hours ten minutes against nine minutes twenty seconds.
PagerDuty wakes you up. Oppex helps you fix it.
What changes for your on-call engineers
Fewer pages. Duplicate alerts are collapsed and related alerts grouped by service, time, labels and AI similarity, with the reason for each grouping shown. Flapping and self-clearing alerts are suppressed, and you can test any routing rule against the last seven days before it goes live.
A head start on the cause. Oppex Nova checks deploys, metrics, logs, traces, past incidents and runbooks, and returns the likely cause, a confidence level and a recommended fix in about two minutes. Approve the fix in Slack, or let automated runbooks run where your policy allows.
No status-update homework. Executives get plain-language updates automatically — what is affected, who owns it, when the next update arrives — on severity change, status change, every 30 minutes and on resolve.
Work from where they already are. Acknowledge, add notes and resolve from Slack, Claude Code, Codex or Cursor, with everything on one incident timeline.
And your engineers stay in control. Oppex proposes; your engineers decide. Every finding links to the signal behind it, low-confidence findings are labelled as such, and nothing runs automatically on a guess.
Oppex customers report 5× faster MTTR, 70% less noise, 50% less on-call effort, 2× fewer late-night calls and 3× faster onboarding. Your own numbers are measured in a free one-week noise audit before you commit.
What it costs
With PagerDuty, noise reduction, AI, runbook automation and status pages are sold as add-ons on top of the per-user plan, or in a separate AI platform sold in seat bundles. One Oppex licence covers paging, noise reduction, AI investigation, runbooks, reporting and status pages.
The closest comparison is Oppex Business against PagerDuty's AI platform, since both include AI investigation:
| Team size | Oppex Business | PagerDuty AI platform | You save each year |
|---|---|---|---|
| 10 users | $2,400 | $2,800 | $400 (14%) |
| 20 users | $4,800 | $12,000 | $7,200 (60%) |
| 50 users | $12,000 | $48,000 | $36,000 (75%) |
Annual list prices as of October 2026, from each vendor's public pricing page. Oppex Business is $20 per user per month billed annually; see the full feature-by-feature comparison and Oppex pricing.
How to switch without the risk
You do not have to switch everything at once, or switch anything before you have seen results:
- Week 0 — free noise audit. Send one week of alerts. We show you your noisiest services and agree a success target with you, for example 50% fewer pages and MTTR halved.
- Week 1 — shadow mode. PagerDuty keeps paging as normal. Oppex receives a copy of every alert, groups and investigates it, and pages nobody.
- Weeks 2–3 — side by side. Oppex pages for your pilot services, with PagerDuty still on as a backup.
- Week 4 — your decision. We review pages, noise, MTTA, MTTR and hours saved against the target. If we miss it, you keep PagerDuty exactly as it was.
- After week 4 — migration complete. Oppex becomes your primary incident management platform.
Our team imports your users, teams, services, on-call schedules and escalation policies from PagerDuty, and connects your existing monitoring, chat and ticketing tools.
"We had been deeply integrated with PagerDuty Enterprise for nearly a decade, but we were able to migrate to Oppex in just a few weeks. Oppex made the transition seamless." — Dirk Krummacker, Director, Software Engineering, SessionM (a Capillary company)
Frequently asked questions
How is Oppex different from PagerDuty?
PagerDuty is built to make sure someone is paged. Oppex covers what happens after the page: it groups related alerts into one incident before anyone is woken, investigates deploys, metrics, logs, traces and past incidents to propose the likely cause and fix in about two minutes, and keeps leadership updated automatically.
How much does Oppex cost compared with PagerDuty?
Oppex Business is $20 per user per month, billed annually, and includes noise reduction, AI investigation, automated runbooks and status pages. At list prices (October 2026) that is $12,000 a year for 50 users against $48,000 for PagerDuty's AI platform, a 75% saving.
Do we have to switch off PagerDuty to try Oppex?
No. Oppex first runs in shadow mode behind PagerDuty: it receives a copy of every alert, groups and investigates it, and pages nobody. You compare the results against a success target agreed up front before you move any service.