Top 6 Incident Management and On-Call Platforms for 2026: incident.io vs PagerDuty vs Rootly vs Grafana IRM vs Better Stack vs FireHydrant
Priced per responder, with the Opsgenie shutdown timeline that is driving the whole market, verified against Atlassian's own pages.
Most people shopping this category in 2026 are not shopping. They are migrating, because Atlassian is retiring Opsgenie. The dates that matter, taken from Atlassian's own pages: Opsgenie stopped being available for purchase on 4 June 2025, and end of support is 5 April 2027. There is an earlier cutoff too. If you migrate to Jira Service Management or Compass with Atlassian's tool, Opsgenie shuts off automatically 120 days after your migration date unless you turn it off sooner. That shutdown is permanent and cannot be undone. Opsgenie's REST APIs keep working until 5 April 2027.
That 120-day clock is the part teams get wrong. The published deadline is 2027, so migrations get scheduled casually, and then a migration date set in March quietly ends Opsgenie access in July.
If you are leaving and do not want Jira Service Management or Compass, the short answer: incident.io for most teams, because it is chat-native and publishes every number including the on-call add-on. PagerDuty if your alert sources include tooling nobody else integrates with. Rootly if you want response and on-call priced symmetrically, or if you qualify for its startup programme. Grafana Cloud IRM if you already live in Grafana Cloud. Better Stack if monitoring and on-call should be one bill. FireHydrant if executable runbooks are the priority, with the Freshworks acquisition weighed in.
Grafana OnCall OSS is no longer an option. It entered maintenance mode on 11 March 2025 and was archived on 24 March 2026, with the repository read-only and the Cloud Connection for mobile push, SMS, and phone notifications discontinued on the same date. Anyone planning a free self-hosted paging stack around it needs a different plan.
One pricing pattern runs through this whole page and it decides the bill. On-call is priced separately from incident response almost everywhere, and the packaging differs by vendor. incident.io charges $10 per user on Team and $20 on Pro on top of the base seat. Rootly sells two $20 products. PagerDuty and Better Stack fold scheduling into the seat. FireHydrant includes Signals in Pro but meters it on alert volume. Compare the total per responder for both functions or the comparison is meaningless.
Every price here was read from the vendor's own pricing page in September 2026, and the Opsgenie timeline was verified against Atlassian's Opsgenie pricing page and its what happens when Opsgenie is turned off support article rather than third-party summaries.
Incident tooling consumes alerts; it does not create them. If your pages are noisy, the fix is upstream in the monitoring layer, not in a new paging vendor. That decision is covered separately in the observability platform comparison, which this page deliberately does not repeat.
Quick Comparison
| Platform | Best For | Free Tier | Published Price (per responder/month) | Honest Catch |
|---|---|---|---|---|
| incident.io | Slack or Teams-native response plus on-call in one product | Basic plan, free, single team on-call | $19 Team (monthly) or $15 annual, plus $10 on-call add-on | On-call is an add-on, so the real price is two line items |
| PagerDuty | Large estates, ITSM integration, and the widest integration catalogue | 5 users, 1 schedule, 1 escalation policy | $25 Professional, $49 Business (monthly list) | Most expensive per seat, and AI Actions are one-time allotments |
| Rootly | Teams that want incident response and on-call priced the same and bundled | Not published on the pricing page | $20 Incident Response, $20 On-Call, bundle discount unquantified | Enterprise pricing unpublished and bundle discount is not a number |
| Grafana Cloud IRM | Shops already standardised on Grafana Cloud metrics and alerting | 3 active IRM users per month | From $20 per active IRM user, plus $19/mo platform fee | Grafana OnCall OSS was archived on 24 March 2026 |
| Better Stack | Small teams wanting uptime monitoring and on-call on one bill | 10 monitors and heartbeats, 1 status page, no SMS or calls | $34 per responder monthly, $29 annual | Free tier has no phone or SMS alerting, which is the point of on-call |
| FireHydrant | Structured major-incident process with runbooks and retrospectives | 10 responders, 2 runbooks, 1 status page, 3 integrations | $25 Pro, billed annually | Being acquired by Freshworks, announced December 2025 |
incident.io
- Best For
- Slack or Teams-native response plus on-call in one product
- Free Tier
- Basic plan, free, single team on-call
- Published Price (per responder/month)
- $19 Team (monthly) or $15 annual, plus $10 on-call add-on
- Honest Catch
- On-call is an add-on, so the real price is two line items
PagerDuty
- Best For
- Large estates, ITSM integration, and the widest integration catalogue
- Free Tier
- 5 users, 1 schedule, 1 escalation policy
- Published Price (per responder/month)
- $25 Professional, $49 Business (monthly list)
- Honest Catch
- Most expensive per seat, and AI Actions are one-time allotments
Rootly
- Best For
- Teams that want incident response and on-call priced the same and bundled
- Free Tier
- Not published on the pricing page
- Published Price (per responder/month)
- $20 Incident Response, $20 On-Call, bundle discount unquantified
- Honest Catch
- Enterprise pricing unpublished and bundle discount is not a number
Grafana Cloud IRM
- Best For
- Shops already standardised on Grafana Cloud metrics and alerting
- Free Tier
- 3 active IRM users per month
- Published Price (per responder/month)
- From $20 per active IRM user, plus $19/mo platform fee
- Honest Catch
- Grafana OnCall OSS was archived on 24 March 2026
Better Stack
- Best For
- Small teams wanting uptime monitoring and on-call on one bill
- Free Tier
- 10 monitors and heartbeats, 1 status page, no SMS or calls
- Published Price (per responder/month)
- $34 per responder monthly, $29 annual
- Honest Catch
- Free tier has no phone or SMS alerting, which is the point of on-call
FireHydrant
- Best For
- Structured major-incident process with runbooks and retrospectives
- Free Tier
- 10 responders, 2 runbooks, 1 status page, 3 integrations
- Published Price (per responder/month)
- $25 Pro, billed annually
- Honest Catch
- Being acquired by Freshworks, announced December 2025
incident.io
Best OverallBest for: Teams that run incidents in Slack or Microsoft Teams and want alerting, escalation, response, status page, and postmortems from one vendor without stitching two products together
“incident.io is the strongest default for most teams leaving Opsgenie, because it is the only product here that is both genuinely chat-native and publishes a full price list including the on-call component. The response side is excellent and the on-call side has closed most of its gap with PagerDuty. Read the pricing carefully though: on-call is an add-on, so a Team plan responder actually costs $25 a month, not $15.”
Pros
- Native Slack and Microsoft Teams incident response, where the incident channel, roles, timeline, and status updates are the product rather than a notification integration
- Free Basic plan covers Slack or Teams response, single-team on-call, a status page, and essential automation, which is a real evaluation path
- Publishes every number including the on-call add-on, so total cost is calculable before a sales call
- On-call can be bought standalone at $20 per user per month for teams that only need paging
- Automatic timeline capture and postmortem drafting remove most of the manual write-up burden that kills retrospective discipline
Cons
- On-call is priced separately, so the headline $15 or $19 Team price understates the real cost by $10 per responder per month
- Pro tier's on-call add-on is $20 per user on top of a $25 base, which puts a Pro responder at $45 a month, close to PagerDuty Business
- Deeply tied to Slack or Teams, so it is a poor fit for organisations that do not run incidents in chat
- Enterprise pricing and the 99.99 percent SLA tier are quote-only
Why Chat-Native Matters For Opsgenie Migrations
Opsgenie was an alerting and on-call tool with a chat integration bolted on. Most teams migrating off it discover their actual incident process already lives in a Slack channel somebody opens manually, with a spreadsheet of roles and a document nobody fills in. incident.io's model matches that reality: the channel is created for you, roles are assigned in the channel, the timeline builds itself, and the postmortem is drafted from it. That closes the gap Opsgenie never covered, which is why a like-for-like paging replacement is usually the wrong target.
Basic: free, includes Slack or Teams response, single-team on-call, status page. Team: $19/user/month monthly or $15/user/month annual, on-call add-on $10/user/month. Pro: $25/user/month, on-call add-on $20/user/month. On-call standalone: $20/user/month. Enterprise: custom, 99.99 percent uptime SLA. As published on incident.io's pricing page, September 2026.
PagerDuty
Best for EnterpriseBest for: Large or regulated estates that need the widest integration catalogue, deep ITSM and change-management ties, and an incident workflow engine with conditionals and loops
“PagerDuty remains the reference implementation and the most expensive per seat. Its integration catalogue is larger than anything else here, its Enterprise workflow engine handles conditionals, loops, and delays that competitors do not, and it is the safest answer in a procurement process. What it is not is cheap or chat-native, and its free plan tops out at five users with one schedule and one escalation policy.”
Pros
- Over 750 integrations, which matters when the estate includes monitoring tools nobody else has bothered to support
- Incident Workflows on Enterprise support conditionals, loops, and delays, so complex escalation and communication policies can be encoded rather than documented
- Deep ITSM integrations and multi-year data access on Business and above, which is usually what regulated buyers actually need
- Free plan includes 100 international phone and SMS notifications per month, so the free tier can genuinely page someone
- 16 percent discount for annual billing moves Professional to $21 and Business to $41 per user per month
Cons
- Highest per-seat list price here: $25 per user per month Professional and $49 Business on monthly billing
- Free plan is capped at 5 users, 1 on-call schedule, and 1 escalation policy, which is a demo rather than a working setup for a real rota
- Chat experience is an integration rather than the native surface, so response coordination feels bolted on compared with incident.io or Rootly
- PagerDuty Advance AI Actions are granted as one-time allotments (1,000 on Professional, 5,000 on Business, 20,000 on Enterprise) rather than recurring quotas, which is easy to misread as a monthly entitlement
- Custom incident types and custom fields are gated behind Business and above
The Integration Catalogue Is The Moat
Every vendor here supports Datadog, Prometheus, and a generic webhook. The difference shows up on the twelfth tool: a legacy APM, a mainframe monitor, a bespoke ITSM. PagerDuty's 750-plus integration catalogue is the reason it keeps winning in large estates, and it is a genuine technical advantage rather than marketing. If your shortlist includes PagerDuty at twice the price, the honest test is to list every alert source you have and check coverage on each vendor before comparing the per-seat number.
Free: $0, maximum 5 users, 1 on-call schedule, 1 escalation policy, 100 international phone/SMS notifications per month. Professional: $25/user/month monthly, $21 annual. Business: $49/user/month monthly, $41 annual. Enterprise: custom. As published on PagerDuty's incident management pricing page, September 2026.
Rootly
Best ValueBest for: Teams that want incident response and on-call from one vendor at one flat price per user, and startups that can qualify for the published discount programme
“Rootly prices incident response and on-call identically at $20 per user per month each, which makes the arithmetic unusually easy: $40 per responder for both, before an unquantified bundle discount. Its on-call feature set is the most complete of the newer entrants, including live call routing, gap detection, and on-call pay calculation. The startup programme is the most generous published offer in this category.”
Pros
- Published, symmetric pricing: $20 per user per month for Incident Response Essentials and $20 for On-Call Essentials
- On-call feature depth beyond paging, including schedule gap detection, live call routing, heartbeat monitoring, and on-call pay calculation
- Published startup discount of up to 50 percent for companies under 100 employees, under $50M raised, and under 5 years old, plus a pay-what-you-can model under 25 employees
- Separate AI SRE product line for teams that want automated investigation, sold separately rather than bundled into the base price
Cons
- No free tier documented on the pricing page, though the on-call product page carries a get-started-for-free call to action, which is an inconsistency worth clarifying with sales
- Bundle discount for combining Incident Response, On-Call, and AI SRE is stated but never quantified, so the real combined price is not calculable from published material
- Enterprise pricing is not published for any of the three product lines
- AI SRE pricing is available on request only, which makes the newest part of the product the least transparent
On-Call Pay Calculation Is A Real Differentiator
Most of this category treats the rota as a routing problem. Rootly treats it partly as a payroll problem, calculating on-call compensation from the schedule. That sounds minor until you work somewhere that owes shift allowances for out-of-hours cover and reconciles them by hand every month. If your organisation pays for on-call, this feature is worth more than several of the AI capabilities the whole category is currently marketing.
Incident Response Essentials: $20/user/month. On-Call Essentials: $20/user/month. Enterprise tiers for both: custom. AI SRE: on request. Bundle discounts available but not quantified. Startup pricing: up to 50 percent off for under 100 employees, under $50M raised, under 5 years in business; pay-what-you-can under 25 employees. As published on Rootly's pricing page, September 2026.
Grafana Cloud IRM
Best ValueBest for: Organisations already running Grafana Cloud for metrics, logs, and alerting, who want paging in the same bill and the same alert pipeline
“Grafana Cloud IRM is the right answer for Grafana Cloud shops and close to irrelevant for everybody else. Pricing is per active IRM user, which is narrower than per seat and genuinely cheaper for teams where only a subset carry the pager. The important fact is negative: Grafana OnCall OSS is not an option any more. It entered maintenance mode on 11 March 2025 and was archived on 24 March 2026, with the repository now read-only.”
Pros
- Priced per active IRM user rather than per seat, so people who never take a page do not cost anything
- Free tier covers up to 3 active IRM users per month, which is enough for a very small rota
- Pro starts at $20 per active IRM user with a $19 monthly platform fee that includes the first 3 active users, and volume discounts apply automatically
- Alerts, dashboards, and paging share one system, which removes an entire integration and an entire vendor relationship for Grafana Cloud users
Cons
- Grafana OnCall OSS is archived as of 24 March 2026 and the repository is read-only, so the self-hosted free path no longer exists
- The OSS Cloud Connection that provided mobile push, SMS, and phone notifications was discontinued on the same date, which strands any remaining self-hosted deployment
- Enterprise requires a minimum $25,000 per year commitment, a hard floor most competitors do not impose
- The active-user billing definition is broad: being on a schedule or escalation chain counts, as does acknowledging an alert or editing an incident, so the billable count can exceed the intuitive one
- Weak fit for teams not already on Grafana Cloud, because the integration advantage is the entire value proposition
What Counts As An Active IRM User
Grafana's definition is worth reading before modelling the bill. A user is active in a month if they are included in an on-call schedule or escalation chain, or if they change an alert group or on-call configuration, receive a page, page someone else, or create or edit an incident. Being on a schedule alone is enough. For a team where 12 engineers rotate through a pager, that is 12 active users, not the three who happened to get woken up.
Free: $0, limited to 3 active IRM users per month, community support. Pro: from $20 per active IRM user per month, plus a $19/month platform fee that includes the first 3 active IRM users, 8x5 email support, automatic volume discounts. Enterprise: custom with a minimum $25,000 per year commitment. Grafana OnCall OSS: archived 24 March 2026, no longer available. As published on Grafana's pricing page and OnCall documentation, September 2026.
Better Stack
Best Free OptionBest for: Small and mid-sized teams that want uptime monitoring, status pages, and on-call from one vendor on one bill, without separate contracts for each
“Better Stack is the cheapest credible way to get monitoring and on-call together, and the bundling is the point: on-call scheduling is included at no extra charge for responders, with unlimited SMS, phone calls, and push notifications. The catch is the free tier. It gives 10 monitors, heartbeats, and a status page, but Slack and email alerts only, which means the free plan cannot do the one thing on-call exists for.”
Pros
- Unlimited SMS and phone call alerts on paid plans rather than metered notification credits, which is unusual in this category
- On-call scheduling included for responders at no additional charge rather than sold as an add-on
- Team members who only need telemetry access cost $0, so observability readers do not consume responder licences
- Monitoring, status pages, and incident management in one product removes a vendor and an integration for smaller teams
- 60-day money-back guarantee, which is a genuine low-risk evaluation path
Cons
- Free tier has no SMS or phone call alerting, so it cannot actually wake anyone up
- Responders are $34 per month monthly or $29 annual, which is higher per head than incident.io Team with the on-call add-on
- Monitor capacity is metered: 10 included, then $25 per month for an additional 50 on monthly billing
- Advanced Slack and Teams incident workflows cost an extra $9 per responder licence per month, so the chat-native experience is a paid upgrade
- Incident response depth (runbooks, structured roles, retrospective tooling) is thinner than incident.io, Rootly, or FireHydrant
Unlimited Notifications Versus Notification Credits
Most of this category meters phone and SMS delivery, and it is the line item people forget. PagerDuty's free plan includes 100 international phone and SMS notifications a month, and metered delivery is a normal part of paid plans across the market. Better Stack includes unlimited SMS and phone calls in the responder seat. For a noisy estate with a lot of out-of-hours paging, that difference can outweigh a lower headline per-seat price elsewhere.
Free: 10 monitors and heartbeats, 1 status page, Slack and email alerts, no SMS or phone calls. Responder seat: $34/month monthly, $29/month yearly, includes unlimited SMS, phone calls, push, and email alerts plus on-call scheduling. Team member (telemetry access only): $0. Additional 50 monitors: $25/month monthly, $21 yearly. Advanced Slack/Teams incident workflows: $9 per responder licence per month. 60-day money-back guarantee. As published on Better Stack's pricing page, September 2026.
FireHydrant
Honorable MentionBest for: Teams whose priority is a structured major-incident process, with codified runbooks, service catalogue context, and disciplined retrospectives rather than paging volume
“FireHydrant has always been the most process-serious product in this category, built around runbooks that execute rather than documents that get read afterwards. Its free tier is also the most generous: 10 responders, a public status page, and real functionality. It is ranked last here for one reason that has nothing to do with the product. Freshworks announced an agreement to acquire FireHydrant on 15 December 2025, expected to close in Freshworks' first fiscal quarter of 2026.”
Pros
- Free tier covers up to 10 responders with 2 runbooks, 1 public status page, and 3 integrations, which is a workable small-team setup rather than a demo
- Runbooks are executable automation attached to incident types, not documentation, which is the strongest answer here to inconsistent incident process
- Service catalogue is built in, so incidents carry ownership and dependency context without a separate portal
- Pro is $25 per responder per month billed annually, with unlimited public status pages and the Signals on-call component included
- Retrospective tooling is integrated with the timeline rather than a separate write-up step
Cons
- Acquisition by Freshworks announced 15 December 2025, so product direction, pricing, and packaging are all subject to change as it is folded into Freshservice
- Signals on-call charges on alert volume rather than per user, which is harder to forecast than a flat per-responder price
- Pro pricing is annual-billing only as published, so there is no monthly escape hatch
- Free tier limits runbooks to 2 and integrations to 3, which is the constraint you hit first
- The acquisition positions it inside an ITSM suite, which is a different buyer from the SRE teams that chose it
Runbooks That Execute
The gap between a documented incident process and a followed one is where most post-incident reviews find their real root cause. FireHydrant's runbooks fire automatically on incident conditions: open the channel, assign the commander, notify the right stakeholder group, start the status page update, open the retrospective. That is a meaningfully different bet from tools that make the process easy to follow manually, and it is the reason FireHydrant is still worth evaluating despite the ownership question.
Free: $0, up to 10 responders, 2 runbooks, 1 public status page, 3 integrations. Pro: $25/responder/month billed annually, 5 runbooks, unlimited public status pages, service catalogue, Signals on-call. Enterprise: custom. Signals alerting is charged on alert volume rather than per user. As published on FireHydrant's pricing page, September 2026. Freshworks announced an agreement to acquire FireHydrant on 15 December 2025.
Which One Should You Pick?
| Use Case | Our Recommendation |
|---|---|
| You are migrating off Opsgenie and do not want to move to Jira Service Management or Compass | incident.io or Rootly. Both publish full prices, both cover alerting plus response rather than paging alone, and both are straightforward to run in parallel during cutover. Schedule the migration deliberately: Atlassian ends Opsgenie support on 5 April 2027, and if you migrate to Jira Service Management or Compass, Opsgenie shuts off automatically 120 days after your migration date. |
| Your alert sources include legacy or unusual tooling nobody else integrates with | PagerDuty. The 750-plus integration catalogue is the actual differentiator at the price. List every alert source you have and check coverage on each shortlisted vendor before you compare per-seat numbers, because a missing integration becomes a maintenance project. |
| You are already running Grafana Cloud for metrics, logs, and alerting | Grafana Cloud IRM. Per active IRM user billing from $20, alerts and paging in one pipeline, and no extra vendor. Do not plan around Grafana OnCall OSS: it was archived on 24 March 2026 and the repository is read-only. |
| A small team needs uptime monitoring and on-call, and the budget is one line item | Better Stack. Responders at $29 to $34 a month include unlimited SMS and phone calls plus on-call scheduling, and telemetry-only team members cost nothing. Budget for a paid plan from day one, because the free tier has no SMS or phone alerting. |
| Incidents are handled inconsistently and the process exists only in a document | FireHydrant. Executable runbooks fire the process automatically rather than relying on someone remembering it. Weigh the Freshworks acquisition announced in December 2025 against that benefit before signing a multi-year term. |
| You are a startup under 25 people and need real on-call without a real budget | Rootly's pay-what-you-can programme for companies under 25 employees, or FireHydrant's free tier at up to 10 responders. incident.io's free Basic plan also covers single-team on-call. Avoid PagerDuty Free, which caps at 5 users with one schedule and one escalation policy. |
How we evaluated
The Opsgenie retirement is pushing a large number of teams to re-buy a category they last evaluated years ago, usually under time pressure. This comparison is built to answer the question that migration actually raises: what does a responder cost, for both paging and response, on the tier we will really be on.
Each platform was assessed on the criteria in the comparison table above:
- Best fit: the team shape and existing toolchain each product genuinely serves, not the broadest possible audience.
- Free tier: what the free plan can actually do, including whether it can deliver a phone call or SMS at all. A free tier that only sends Slack messages cannot do on-call.
- Published price per responder: the real cost for both incident response and on-call, since the two are packaged differently by every vendor here.
- Notification economics: whether SMS and phone delivery is unlimited, metered, or charged on alert volume.
- Honest catch: the failure mode, pricing trap, or corporate-ownership risk of each product, stated plainly.
What we verified, and when
Last verified: September 2026. Every price and limit was read from the vendor's own pricing page or documentation this month:
- PagerDuty incident management pricing for the Free plan caps, Professional and Business list and annual prices, and the AI Actions allotments.
- incident.io pricing for the Basic, Team, and Pro prices, the on-call add-on rates, and the standalone on-call price.
- Rootly pricing for the $20 Incident Response and $20 On-Call Essentials prices and the published startup discount criteria.
- Grafana pricing for the IRM free tier, the per active IRM user rate, the platform fee, the Enterprise minimum commit, and the definition of an active IRM user.
- Better Stack pricing for the free tier limits, responder seat prices, monitor overage, and the advanced chat workflow add-on.
- FireHydrant pricing for the free tier limits and the Pro responder price.
The Opsgenie timeline was verified against Atlassian's own material rather than any third-party write-up. The Opsgenie pricing page gives the 4 March 2025 announcement, the 4 June 2025 end of new sales, and the 5 April 2027 end of support. The support article what happens when Opsgenie is turned off gives the 120-day automatic shutdown after migration and the API cutoff. The Grafana OnCall OSS archival dates come from Grafana's own maintenance mode notice.
The FireHydrant acquisition is from Freshworks' investor relations announcement dated 15 December 2025.
What we could not verify
Enterprise pricing is unpublished for every vendor here, so the top of each range is not comparable. Rootly's bundle discount for combining Incident Response, On-Call, and AI SRE is stated on its pricing page but never quantified. Rootly's pricing page shows no free tier while its on-call product page carries a get-started-for-free call to action, and that inconsistency could not be resolved from published material. Whether the Freshworks acquisition of FireHydrant has formally closed could not be confirmed from a Freshworks or FireHydrant page this session.
Editorial independence: this is a vendor-neutral comparison with no paid placements, sponsorships, or affiliate links. Rankings reflect fit for the stated use cases, not commercial relationships. No product here was tested hands-on; the evaluation rests on vendor documentation, published pricing, and vendor announcements, all cited above.
Frequently Asked Questions
When exactly does Opsgenie shut down, and what is the migration deadline?
What is the cheapest real replacement for Opsgenie?
Is Grafana OnCall still free and self-hostable?
Is FireHydrant still safe to buy after the Freshworks acquisition?
Do I need separate incident management and on-call tools?
Which incident management platform has the best free tier?
How does this relate to our observability tooling?
Related Comparisons
Platform Engineering
Top 6 Internal Developer Portals for 2026: Backstage vs Port vs Cortex vs OpsLevel vs Spotify Portal vs Humanitec
6 tools compared
Workflow Orchestration
Top 6 Workflow Orchestration Platforms for 2026: Temporal vs Airflow vs Prefect vs Dagster vs Kestra vs Windmill
6 tools compared
Container Infrastructure
Top 8 Container Registries for 2026: Pull Rate Limits, Retention, Scanning, and Egress Priced
8 tools compared
Backend Platform
Top 7 Supabase Alternatives for Startups (2026)
7 tools compared