How to use Atlassian Intelligence link similar work items in 2025

If you’ve spent any time managing a Jira instance, you know the drill. A new ticket lands in the queue, and you get that nagging feeling you’ve seen it before. Like, last week. And the week before that. Dealing with the same issues over and over can feel like you're stuck on a hamster wheel. Finding related tickets is a good first step, but let’s be honest, the real goal is to get them solved and off your plate for good.

Atlassian Intelligence has a feature that tries to help with this, but it’s really just scratching the surface. This guide will walk you through what the "link similar work items" feature actually does, where it’s genuinely useful, where it falls short, and how you can go from just linking issues to actually getting them resolved automatically.

What is the Atlassian Intelligence link similar work items feature?

At its core, the Atlassian Intelligence link similar work items feature is like a little helper for your agents. It uses natural language processing (NLP) to read the titles and descriptions of your Jira issues and then suggests others that look similar. This is all part of Atlassian's bigger AI push, which they've recently started calling Rovo.

The main idea is to help your team find and connect related tickets, Confluence pages, and other documents right from the ticket they're working on. For example, if you're looking at a live incident, it might pop up a little panel with "Similar incidents" or give you a "Find resources" button to dig up old post-incident reviews. The whole point is to give your team context without making them search for it manually.

It’s important to be clear about what this feature is, and what it isn't. It’s all about showing information to a person. It gives you hints and suggestions, but it doesn't actually do anything to the ticket itself. The agent still has to read through everything, figure out what’s relevant, and decide what to do next.

Where the Atlassian Intelligence link similar work items feature helps in Jira

Even with its limitations, this feature can be pretty handy for teams trying to get a handle on their internal workflows.

Getting agents up to speed

For new agents, or even experienced ones dealing with a weird new problem, this feature can be a real time-saver. If a similar bug has popped up before, they can instantly see how it was fixed, who was involved, and what happened.

For instance, an agent gets a ticket titled "login button not working." The AI points them to three older incidents with almost identical descriptions. The agent clicks through, sees it’s a recurring server issue that needs a specific reboot sequence, and avoids wasting an hour on diagnostics.

Spotting duplicate tickets and trends

One of the biggest headaches in any support queue is when ten different people report the exact same problem. This feature helps you catch those duplicates almost right away. By linking them all to a main ticket, you can keep all your communication in one place and stop running in circles.

Picture this: a small service outage leads to a flood of 50 tickets all screaming about the same error message. An agent can use the AI suggestions to link them all to a single master incident. Now, you only need to post updates in one spot, keeping everyone informed without the chaos.

Making incident resolution faster

When you're in the middle of a live incident, every second counts. Digging up old post-incident reviews (PIRs) or knowledge base articles is crucial. The "Find resources" function, powered by Atlassian Intelligence, can pull these up right when you’re feeling the pressure.

Let's say a critical database starts acting up. The AI might surface a PIR from six months ago describing a nearly identical event. That document could have the exact fix and mention the engineers who solved it, giving the person on-call a huge head start.

Where the Atlassian Intelligence link similar work items feature falls short

Okay, let's get practical. The feature is helpful, but it has a natural ceiling because it's built for assistance, not full-on automation.

It's reactive, not proactive

The feature only jumps into action after a ticket has been created and an agent is looking at it. It does nothing to stop that ticket from being created in the first place, and it doesn't solve it before an agent has to spend time on it.

This means your team is still opening, reading, and triaging every single ticket that comes in, even if the fix is simple and well-documented. It helps with the how of solving the problem, but it doesn’t do much to reduce the initial workload.

It still relies on manual work

The AI gives you suggestions, but a human has to do the rest. An agent still needs to read the linked items, decide if they’re actually a match, piece together the information, and then write a response or take action. The mental effort doesn't disappear; it just shifts from searching to analyzing.

As any seasoned Jira admin will tell you, they spend most of their day putting out fires. This feature is another tool for firefighting, but it doesn't take those tasks off their plate. That leaves less time for the proactive stuff, like improving workflows or building better documentation.

It's stuck in the Atlassian ecosystem

Atlassian Intelligence is at its best when it's looking at data inside its own world, mostly Jira and Confluence. But let’s be real, important information is scattered all over the place: in shared Google Docs, old Slack messages, internal wikis, or external help centers.

If the answer to a ticket is buried in a Google Doc that details a specific customer's setup, your agent won't see it suggested in Jira. They'll have to go back to searching the old-fashioned way, which kind of defeats the purpose of having an AI assistant.

Atlassian Intelligence pricing for Jira

It's also worth pointing out that these AI features don't come with Atlassian's Free plan. They're part of the paid tiers, and what you get depends on how much you pay.

Here’s a quick look at the plans that include Rovo (Atlassian Intelligence) features:

Feature	Standard Plan	Premium Plan	Enterprise Plan
Price	$7.91 / user / month	$14.54 / user / month	Contact Sales (Annual Billing)
Rovo AI Features	Included	Included	Included
AI Credits	25 per user/month	70 per user/month	150 per user/month
Indexed Objects	100 per user	250 per user	625 per user
Key Use Case	Basic AI search, content generation, and automation.	Advanced planning, more automation, and higher AI limits.	Enterprise-grade scale, security, and unlimited automation.

The cost can creep up as your team gets bigger, and the new AI credit system means you have to watch your usage. This can make your budget feel a bit unpredictable, especially when things get busy.

A better way: Moving from linking items to resolving them

Linking similar items is a nice start, but the end game for any team is to cut down on manual work and solve problems instantly. That requires an AI that can not only find information but also understand and act on it.

This is where a platform like eesel AI comes into the picture. It’s built to be that next step, adding a layer of true automation on top of tools you already use, like Jira Service Management.

Bringing all your knowledge together

Unlike tools that are walled off inside the Atlassian ecosystem, eesel AI connects to all of your company knowledge, no matter where it is. It hooks into Confluence, of course, but it also connects to Google Docs, past tickets, Slack threads, Notion pages, and more.

This gives your AI agent a complete picture of your company's brain. It can pull answers from any connected source to resolve tickets, making sure nothing gets missed.

Automating actions, not just suggesting links

The eesel AI Agent doesn't just point to a related ticket and call it a day. You can actually set it up to take direct action on its own. This could be anything from drafting and sending a reply in your company's voice, to correctly tagging a ticket, looking up order info in Shopify with an API call, and closing it out, all without a person ever getting involved.

This directly tackles the "it's still manual work" problem. Simple, repetitive tickets get handled automatically, 24/7. That frees up your skilled agents and admins to focus on the tricky, high-value problems that really need a human brain.

Simulating and deploying without the stress

Letting an AI loose on your support queue sounds a little scary, right? That’s why eesel AI has a simulation mode. You can test your AI agent on thousands of your past tickets in a safe environment. You get to see exactly how it would have replied, get a clear idea of its resolution rate, and spot any holes in your knowledge base.

This takes all the guesswork out of deployment. You can tweak the AI's behavior, set specific rules for what kinds of tickets it can handle, and roll it out one step at a time. Start with one type of ticket, see how it does, and grow its responsibilities as you get more comfortable. It's a level of control that a simple on/off feature can't give you.

Atlassian Intelligence link similar work items is good, resolving is better

The Atlassian Intelligence link similar work items feature is a solid tool. It gives agents helpful context, makes it easier to spot duplicate issues, and can speed up incident resolution. It’s a definite step toward a more organized workflow.

But its value hits a ceiling because it's reactive and still depends entirely on your team to do the actual work. To really change how your support and ITSM teams operate, the goal has to shift from helping agents to automating resolutions.

Platforms like eesel AI are the next step in that evolution. They connect all of your scattered knowledge to handle tickets from start to finish on their own. By doing that, they give your team back the time they need for the strategic work that actually pushes the business forward. If you’re ready to go beyond just linking tickets and start resolving them automatically, it’s worth seeing what a dedicated AI support platform can do.