A guide to using Atlassian intelligence to create incidents from alerts

If you’ve ever been on call in an IT ops team, you know the drill. The alerts just don't stop. Notifications ping in from every monitoring tool you own, creating a constant stream of noise that makes it ridiculously hard to spot a real fire. Trying to manually create an incident for every potential issue is a slow, tedious grind of copy-pasting details into a ticket. It’s a recipe for burnout and, worse, it means critical incidents take longer to address.

This is exactly the problem that AI automation promises to solve. The idea is to have a smarter, faster way to handle incident response so your team can focus on fixing things. Atlassian has its own take on this with Atlassian Intelligence, which is designed to help teams bring some order to the chaos of alert management.

In this guide, we’ll take a practical look at how Atlassian Intelligence handles creating incidents from alerts. We'll walk through what it does, what it does well, and, just as importantly, some of the real-world limitations you’ll likely hit.

How does Atlassian Intelligence create incidents from alerts?

Atlassian Intelligence, which you might also hear called Rovo, is a layer of AI features built into Atlassian’s products, including the widely used Jira Service Management (JSM). When it comes to managing incidents, its main purpose is to analyze the flood of incoming alerts from your monitoring tools, spot patterns, and help you escalate actual problems into formal incidents more easily.

You can think of it as a smart filter for your on-call team. Instead of someone having to manually read through dozens of similar alerts to figure out what’s going on, the AI groups them together. From there, it helps create an incident by suggesting a title, description, and even a priority level based on what it sees in the alert data.

The whole goal is to reduce manual work and create more consistent incidents. By speeding up that initial detection and triage phase, you can hopefully shorten the entire incident lifecycle and bring down your Mean Time to Resolution (MTTR).

Key features for creating incidents from alerts

So, what does this look like in practice? Atlassian’s system for turning alerts into incidents relies on a few core capabilities, from smart grouping to configurable automation rules.

AI-powered grouping and summarization

One of its most immediately useful features is how the AI automatically groups similar alerts. This is a direct answer to the classic problem of "alert fatigue," where engineers get so many notifications that they eventually start tuning them out, including the important ones.

The AI scans for similarities in alert titles, descriptions, and other data. For instance, if your database is having a bad day and sends out five slightly different "high latency" alerts from various nodes, the AI is smart enough to see they're all related to the same event. It bundles them into a single group and then gives you a quick summary so your team can understand the situation at a glance without reading every single line. It’s a pretty effective way to cut through the noise.

Semi-automated incident creation

For the most part, creating an incident with Atlassian Intelligence isn't a fully hands-off process. It’s designed to have a person involved. An on-call engineer will see an alert (or a group of them) that seems serious and will kick things off by clicking "Create Incident."

This is where the AI acts as a helpful assistant. It pre-populates the new incident form with a suggested title, a detailed description, and a priority level, all pulled from the alert's content. The key thing to remember is that it's not fully autonomous. The engineer still needs to review the AI's suggestions, make any tweaks, and then confirm to officially create the incident. It’s more of a copilot that makes the process faster, not a system that runs the show on its own.

Automation rules for hands-off creation

If what you're really after is true, no-touch automation, you'll need to get your hands a little dirty with some extra setup. This is done by configuring either Automation for Jira or Incident Rules, a feature that was originally part of Opsgenie.

These tools run on straightforward "if this, then that" logic. A Jira admin can set up a rule like, "If an alert comes in with 'Priority: Critical' AND the source is our main production database, then automatically create a new incident." When those specific conditions are met, an incident gets created and assigned without anyone having to do anything. This is great for predictable, high-stakes failure scenarios, but it also brings up a key point: this level of automation isn't on by default. It takes some technical skill to build and maintain these rules.

Limitations and challenges of using Atlassian Intelligence

While Atlassian's AI is a decent step in the right direction, it has some pretty big limitations, especially for teams that rely on more than just Atlassian tools to get their work done.

Confined to the Atlassian ecosystem

The biggest hurdle is that Atlassian Intelligence lives entirely within the Atlassian world. It works great if all your runbooks, documentation, and procedures are stored in Atlassian products like JSM and Confluence.

But let's be honest, that’s not how most companies operate. Your most important incident runbook might be a Google Doc that everyone collaborates on. Your team’s standard operating procedures could be neatly organized in Notion, and critical customer information might be in a separate help desk like Zendesk. Atlassian's AI has no visibility into any of that external knowledge, so it can't use it to create a more detailed, context-rich incident.

This is where a tool like eesel AI takes a different approach. It’s built to be an intelligence layer that connects to all your different knowledge sources (over 100 of them). It can pull information from Google Docs, wikis, and other apps to give its AI a complete picture of the situation, leading to much smarter automation.

An infographic illustrating how eesel AI connects with various external knowledge sources, a key differentiator from the siloed Atlassian Intelligence.

Complexity in setup and configuration

As mentioned earlier, getting to a place of full automation requires more than just checking a box. It depends on a Jira administrator who can build, test, and maintain a library of automation rules. As your services and monitoring tools change over time, these rules can get complicated and fragile, becoming a maintenance headache.

This reliance on a technical admin can become a real bottleneck. By contrast, eesel AI is designed so that anyone can set it up. You can connect your help desk and knowledge sources with simple, one-click integrations and have powerful AI workflows up and running in minutes. There's no need to file a ticket and wait for a developer or Jira admin to write custom rules for you.

Limited flexibility for custom actions

Atlassian’s automation is good at doing things inside the Atlassian suite, like creating tickets, adding comments, or changing an issue's status.

But what about when resolving an incident requires talking to an external system? Maybe you need to look up an order status in Shopify or check customer details in an internal database. With the native tools, this usually means someone has to build a custom integration or struggle with webhooks, which can be clunky.

eesel AI handles this with a customizable workflow engine that includes custom API actions. You can easily teach your AI agent how to look up real-time information from any third-party system that has an API. This means it can do more than just create incidents; it can actively help resolve them by gathering the data your team needs, right when they need it.

Jira Service Management pricing

It's also important to talk about the cost. Atlassian Intelligence features aren't part of every plan. To use the AI-powered alert grouping and incident creation, you need to be on a Premium or Enterprise plan for Jira Service Management. These features are not included in the Free or Standard tiers.

Since Opsgenie is no longer sold separately and its features have been folded into JSM, your JSM plan is what dictates your access to these AI capabilities.

Plan Tier	AI for Incident Creation	Best For
Free	Not Available	Small teams just getting started with basic ticketing.
Standard	Not Available	Growing teams that need core ITSM features but not advanced AI.
Premium	Included (Alert Grouping, AI-assisted Incident Creation)	Teams that need robust, AI-powered incident management.
Enterprise	Included (All Premium features + advanced controls)	Large organizations with complex security and governance needs.

This pricing structure can be a tough pill to swallow for teams that want to use AI but aren't ready for a big subscription upgrade. For comparison, eesel AI offers straightforward and predictable pricing. All the core AI products, including the autonomous AI Agent and AI Triage tools, are included in every plan. Costs are based on AI usage, not on which features you want to unlock, making it a more accessible option for a wider range of teams.

A more flexible approach to automated incident management

When you put it all together, the challenges of a platform-native AI become pretty clear: your company’s knowledge gets stuck in silos, the setup can get complicated, and the automation often isn't as flexible as modern teams require.

This is why eesel AI was created, to offer a more powerful and adaptable alternative that works with the tools you already have, including JSM. It's meant to be an intelligence layer that sits on top of your existing tech stack, not a replacement that forces you into a big migration project.

With eesel AI, you can:

• Connect all your knowledge: Link JSM with Confluence, Google Docs, Slack conversations, and even past tickets from any help desk to give your AI a single source of truth.

• Get up and running in minutes: The platform is truly self-serve. Thanks to one-click integrations, you can start seeing results on the first day instead of waiting months for a complex implementation to finish.

• Test everything with confidence: Use the simulation mode to see exactly how the AI would have handled thousands of your past alerts and tickets. This lets you fine-tune its behavior and get an accurate idea of its performance before you ever activate it for your team.

The eesel AI simulation dashboard shows how the AI would have handled past incidents, allowing teams to test and build confidence before activation.

Streamlining incident response with Atlassian Intelligence

Atlassian Intelligence offers a solid entry point for teams that are deeply committed to the Atlassian ecosystem and want to explore automated incident creation. It can definitely help cut down on noise and speed up some manual steps.

However, for most modern IT and support teams who use a mix of different tools, its walled-garden approach creates some serious constraints. The future of effective incident management isn't about locking your team into a single vendor's ecosystem; it's about connecting knowledge and workflows across all the systems you already know and love.

Get started with effortless ITSM automation

Ready to connect all your knowledge sources and build an incident response workflow that’s actually intelligent?

Try eesel AI for free and see how you can automate ITSM tasks across your entire tool stack in just a few minutes.