You cannot buy Opsgenie in 2026 at any price. Atlassian stopped new purchases and trials on June 4, 2025, and support ends on April 5, 2027, at which point access is shut off and un-migrated data is deleted. Existing customers renewing before that date pay the published list rate: $9.45 per user per month on Essentials, $19.95 on Standard and $31.90 on Enterprise when billed annually, with a free tier for up to 5 users. So the useful question is no longer what Opsgenie costs. It is what your replacement costs, and how many days you have left.
From September 2026 that runway is 215 days. Seven months sounds comfortable until you remember what an on-call migration involves: rebuilding escalation policies, re-pointing every integration that fires alerts into Opsgenie, re-teaching the rotation, and proving the new system pages the right person at 3am before you switch the old one off. Teams that treat this as a renewal decision rather than a project tend to discover the integration inventory late.
Opsgenie pricing 2026: the full rate card
These are the list prices published on Atlassian's own Opsgenie pricing page, read in September 2026. All figures are US dollars per user per month. They apply to existing customers only, since new purchases closed in June 2025.
| Plan | Annual billing | Monthly billing | Annual discount | What it adds |
|---|---|---|---|---|
| Free | $0 | $0 | n/a | Up to 5 users, 100 SMS account wide, no voice notifications, 1 routing rule |
| Essentials | $9.45 | $11.55 | 18.18% | Unlimited alerting, on-call scheduling and escalations |
| Standard | $19.95 | $24.15 | 17.39% | Incident command, richer routing and reporting |
| Enterprise | $31.90 | $38.50 | 17.14% | Advanced compliance, service dependencies, support commitments |
The annual discount gets worse as you go up
This is a small thing nobody writes about, and it runs against the usual instinct. Vendors normally apply one flat annual discount across the range. PagerDuty, for example, takes exactly 16% off both of its paid tiers. Opsgenie does not: the break is 18.18% on Essentials, 17.39% on Standard and 17.14% on Enterprise. The cheapest plan gets the biggest proportional reward for paying up front, and the most expensive plan gets the smallest.
In absolute dollars it still runs the other way, because the percentages are applied to bigger numbers. On a 25 person team, choosing monthly billing over annual costs an extra $630 a year on Essentials, $1,260 on Standard and $1,980 on Enterprise. If you are inside the final renewal window anyway, monthly billing buys flexibility during a migration that may well be worth paying for. That is the one scenario where the worse rate is the right call.
What Opsgenie actually costs a real team
Worked bills at list, annual billing, so you can find your own headcount.
| Users | Essentials per year | Standard per year | Enterprise per year |
|---|---|---|---|
| 5 (free tier limit) | $567.00 | $1,197.00 | $1,914.00 |
| 10 | $1,134.00 | $2,394.00 | $3,828.00 |
| 25 | $2,835.00 | $5,985.00 | $9,570.00 |
| 50 | $5,670.00 | $11,970.00 | $19,140.00 |
The free plan caps at 5 users, so the sixth person on the rotation is the expensive one. Adding them converts the whole account, and the smallest paid outcome is 6 users on Essentials at $680.40 a year, up from nothing. This is the same cliff PagerDuty has at exactly the same headcount, and it catches the same teams: a five person startup runs free for two years, hires one SRE, and picks up a four figure annual line item that nobody budgeted.
The volume ladder is steeper than most
Opsgenie's published Essentials rate falls from $9.45 per user in the 1 to 25 band to $5.25 in the 501 to 1,000 band, a 44.4% cut, and to $3.70 above 10,000 users, a 60.8% cut off list. That is a real enterprise discount curve rather than a token one. It is also, in 2026, close to academic: an organization large enough to reach those bands is exactly the kind of customer Atlassian is steering into Jira Service Management, and the sales conversation will be about the destination rather than the ladder.
Opsgenie does not monitor anything
Worth settling before you price a replacement, because it changes what you are shopping for. Opsgenie is an alerting and on-call platform. It receives an event from something else, decides who is on call, escalates when that person does not acknowledge, and records what happened. None of those jobs detects an outage. Every alert in your Opsgenie account was generated by a monitoring tool sitting upstream of it.
So if you are pricing the end of Opsgenie, you are pricing one of two products. The paging layer is the one being retired. The detection layer, whatever is actually watching your endpoints, is untouched, and this is a reasonable moment to check whether that layer is any good. A paging tool cannot page you about an outage nobody noticed, and a five minute check interval is the most common reason nobody noticed. We worked through what interval each availability target actually requires in how often you should check uptime.
What replaces it, and what that costs
Atlassian's recommended path is Jira Service Management, which absorbed the on-call and alerting functionality. Atlassian does not publish a clean side by side rate against the old Opsgenie tiers, and third party pricing trackers in 2026 report Jira Service Management Cloud at roughly $20 per agent per month on Standard and roughly $51 per agent per month on Premium, with a free tier for up to 3 agents. Treat those as reported figures rather than quoted ones and confirm your own number with Atlassian, because agent counts and existing Atlassian entitlements move the total a long way.
On the reported Standard figure the swap looks like a wash. Ten seats of Opsgenie Standard was $2,394 a year; ten Jira Service Management Standard agents at $20 is $2,400, a difference of six dollars. That symmetry is why a lot of teams assume the migration is free. Two things break the illusion. Premium is not a wash at all, running roughly 2.6 times the Standard rate. And the billing unit changes from user to agent, and those are not the same population: an on-call engineer who only ever acknowledged pages may or may not need a full agent seat depending on how you configure it. Count the seats you will actually license before you assume parity.
The deadline is a data deadline, not just a support deadline
The April 5, 2027 date is worth reading carefully. Atlassian's own wording is that access to Opsgenie will be shut off, the product will no longer be accessible, and any un-migrated customer data in Opsgenie will be deleted. That is not a grace period with degraded support. It is deletion. Post-incident timelines, escalation history and the audit trail of who acknowledged what all live in that data, and for regulated teams that history is often something you are required to retain. A vendor end-of-life with a hard deletion date is really an obligation with a due date, and it belongs in whatever register your team uses to track compliance obligations and their deadlines rather than in one engineer's calendar.
One earlier deadline is easy to miss: teams whose Opsgenie came bundled with Jira Service Management reportedly lost access in October 2025, well ahead of the standalone date. If your Opsgenie arrived as part of a JSM subscription rather than as a separate purchase, check which timeline you are on.
Opsgenie pricing questions
Can you still buy Opsgenie?
No. Atlassian ended new purchases and trials on June 4, 2025. Existing customers can still add seats and renew, but only up to the April 5, 2027 end of support date. If you are evaluating Opsgenie for the first time in 2026, it is not available to you at any price and you should be comparing its successors instead.
How much is Opsgenie per user?
At list, $9.45 per user per month on Essentials, $19.95 on Standard and $31.90 on Enterprise when billed annually. Billed monthly those become $11.55, $24.15 and $38.50. The free plan covers up to 5 users with 100 SMS notifications account wide, no voice notifications and a single routing rule.
Is Opsgenie free?
There is a free tier for up to 5 users, and it is genuinely usable for a small rotation. The limits that bite are the account wide allowance of 100 SMS notifications a month, the absence of voice notifications entirely, and a maximum of one routing rule, which means every alert reaches the same place regardless of what generated it.
What is the difference between Opsgenie Essentials and Standard?
Essentials covers the core job: unlimited alerting, on-call schedules and escalation policies. Standard roughly doubles the price and adds incident command features, richer routing logic and more reporting. For a single team running one rotation, Essentials is usually enough. Standard earns its keep when several teams share an alerting estate and you need per-service routing.
What happens to my data on April 5, 2027?
Atlassian states that access is shut off, the product becomes inaccessible, and un-migrated customer data is deleted. Export anything you need to retain, including incident history and escalation records, well before the date rather than in the final week.
Is PagerDuty cheaper than Opsgenie?
Not at the entry level. Opsgenie Essentials at $9.45 per user annually undercuts PagerDuty Professional at $21. Higher up they converge: Opsgenie Standard at $19.95 sits just under PagerDuty Professional, and Opsgenie Enterprise at $31.90 sits between PagerDuty's two paid tiers. Both cap their free plans at 5 users, and neither detects anything on its own. The full PagerDuty rate card is in how much PagerDuty costs.
What to do with the 215 days
Inventory first. List every integration that currently fires an alert into Opsgenie, because that list is the actual scope of the migration and it is almost always longer than people expect. Then decide whether you are moving the paging layer only, or using the forced move to reconsider the detection layer underneath it, which is the part that determines whether you get paged at all. If your monitoring runs on five minute checks, no paging tool you migrate to will fix a 99.99% target you cannot measure. Our own flat pricing is on the uptime monitoring pricing page, and if you run in Azure, the check interval question has a specific and dated answer right now on Azure uptime monitoring, because Application Insights URL ping tests retire on September 30, 2026.
Whatever you pick, do the cutover with both systems live for a fortnight and run a deliberate test page through the new path. The failure mode nobody plans for is not the migration going wrong loudly. It is the new escalation policy quietly not paging anyone, and finding out during a real incident.