A complete guide to Atlassian Intelligence automations via natural language

Atlassian is rolling out a new set of AI features called Atlassian Intelligence, and one of the most interesting bits is the ability to create automations using plain English. The idea is pretty cool: you just describe a workflow you want, and the AI is supposed to build the automation rule for you in Jira or Confluence.

It sounds great for teams who want to automate tedious tasks without having to wrestle with complicated rule builders. But how well does it actually work when you put it to the test?

This guide will give you an honest look at the feature. We'll dig into what it does well, how to get it set up, where its limits are, and how it compares to AI automation tools that can connect to all your company's knowledge, not just what's in Atlassian.

What is Atlassian Intelligence?

First off, Atlassian Intelligence isn't something you can buy on its own. It's a collection of AI features that are baked directly into Atlassian’s cloud products, powered by a mix of OpenAI’s tech and Atlassian's own language models. You can think of it as an add-on that’s only available on their pricier Premium and Enterprise plans.

The main attraction for a lot of teams is the "natural language for automation" tool. Its whole job is to take a simple text prompt, like "when a new bug is reported, assign it to the lead dev," and turn it into a formal "if this, then that" automation rule. This lets you skip the fiddly process of picking triggers, conditions, and actions from endless dropdown menus, which can definitely save some time and make automation more accessible for everyone on the team.

Key features and use cases

The process itself is pretty simple. You type a command, the AI suggests a rule, and then you can look it over, tweak it if needed, and set it live. It's a handy way to get routine tasks off your plate across the Atlassian suite.

Automating workflows in Jira Service Management

If you work in support, you know how much pressure there is to be fast and consistent. Manually escalating tickets or trying to remember every SLA is a recipe for mistakes and wasted time. With natural language, you can set up rules to handle this for you.

A classic example is managing SLA breaches. You could write a prompt like: "When a ticket's SLA is about to breach, change its status to Escalated and ping the on-call channel in Slack." This helps your team jump on urgent issues without having to manually watch the queue all day.

Streamlining content management in Confluence

Let's be real, a company knowledge base can easily turn into a mess of outdated articles. Keeping it organized is a manual job that almost always falls to the bottom of the to-do list. Natural language automation can help with some basic housekeeping.

For example, you could set up a prompt like this: "Every six months, find pages that haven't been updated in a while, archive them, and shoot an email to the person who wrote them." A simple rule like this keeps your documentation from getting stale and makes sure your team can actually trust the info they find.

Managing development tasks in Jira Software

Agile teams live and die by their processes. When someone forgets to create a sub-task or assign work to the right person, it can throw a wrench in the whole sprint. You can use AI to help enforce these processes without nagging anyone.

Think about a prompt for your development workflow: "When an issue gets moved to 'In Review', create a sub-task for QA testing and assign it to our QA lead." This makes sure every bit of work goes through the right quality checks, making your development cycle a lot more predictable.

Setup, requirements, and pricing

Getting started with Atlassian Intelligence isn't as simple as just flipping a switch for your team. There are a few hoops to jump through.

How to enable Atlassian Intelligence

Activating this feature is a decision that has to be made at the org level, so you'll need an admin to get it done.

1. An organization admin has to log in at "admin.atlassian.com".

2. From there, they'll need to navigate to the Atlassian Intelligence section in the settings.

3. Finally, they can choose to activate the features for products like Jira or Confluence.

The takeaway here is that individual teams can't just spin this up on their own. It has to be turned on for the entire organization, which might not work for every company's budget or security policies.

Plan requirements and pricing

Here's the biggest catch: Atlassian Intelligence is not included in all plans. It's bundled exclusively with the Premium and Enterprise versions of Atlassian’s cloud products.

To give you an idea, Jira Premium starts at roughly $16 per user, per month. This means you're not just paying for an AI feature; you're upgrading to a full suite of premium tools you might not even need. This pricing can be a dealbreaker for smaller teams or businesses that justwant powerful AI without committing to an expensive, company-wide license.

Atlassian Intelligence: A contrast with self-serve platforms

This is a pretty different model from what you see with dedicated AI platforms. Tools like eesel AI are designed for a quick, self-serve setup. You can connect your help desk and knowledge sources in a few minutes and get started on your own. The pricing is usually more transparent and based on usage, so you aren’t locked into a costly subscription for a bunch of extra features. You can get up and running in minutes, not months, without having to talk to a salesperson.

Key limitations and a flexible alternative

While having a built-in tool is nice, Atlassian's native AI comes with some pretty big limitations that might be a problem for teams with bigger automation plans.

Limited component support and custom actions

Right away, you'll notice Atlassian Intelligence can't handle a lot of common automation tasks. The official documentation points out that it doesn't support several key components, including:

• Triggers from incoming webhooks

• Actions that "send a web request" (which is how you'd call external APIs)

• Actions to "delete an issue"

So, what does that actually mean for your team? It means you can't build automations that talk to your other tools. Need to pull customer info from your CRM, check an order status in Shopify, or kick off a workflow in another app? You can't. Your automations are stuck inside the Atlassian bubble.

Constrained and siloed knowledge sources

The AI only knows what's inside your Jira and Confluence instances. But let’s be honest, that’s just a fraction of where your company’s important information lives.

You’ve got knowledge spread out everywhere: in Google Docs, Notion pages, Slack threads, and old support tickets in help desks like Zendesk or Freshdesk. Atlassian's AI can't access any of that, which means its automations are always working with an incomplete picture.

This infographic shows how eesel AI connects to over 100 knowledge sources, unlike the siloed approach of Atlassian Intelligence.

The alternative: A unified and customizable AI platform

This is where a platform like eesel AI is built to solve these exact problems.

• Bring all your knowledge together: eesel AI connects to over 100 different sources, including your help desk, wikis, and chat tools. This gives the AI a complete view of your company knowledge, so its automations are far more accurate and aware of the full context.

• Get full control: The workflow engine in eesel AI is completely customizable and supports custom API actions. This is a direct answer to Atlassian's limits, letting you build automations that can fetch data from other systems, trigger webhooks, and perform actions wherever you need them.

• Test with confidence: Before you let an AI loose on your customers, you have to know it works. eesel AI has a simulation mode that lets you test your automations against thousands of your past tickets. You can see exactly how it would have performed and get solid forecasts on resolution rates before you go live.

The eesel AI simulation dashboard allows testing automations on historical data, a key feature for platforms using Atlassian Intelligence Automations via Natural Language.

Here’s a quick breakdown of how the two stack up:

Feature	Atlassian Intelligence Automation	eesel AI
Knowledge Sources	Limited to Jira & Confluence	100+ sources (help desks, wikis, chat, etc.)
Custom Actions	Very limited (no webhooks or external API calls)	Fully supported via customizable API actions
Setup Time	Needs admin activation & a Premium plan	Self-serve, live in minutes
Testing & Rollout	Manual rule inspection	Powerful simulation on historical data
Pricing Model	Bundled with expensive per-user plans	Transparent, interaction-based plans

Is Atlassian Intelligence automation enough for your needs?

Atlassian's natural language automation is a decent feature for teams that are already paying for Premium or Enterprise plans and only need to automate simple tasks inside Jira and Confluence. For basic, internal workflows, it’s convenient.

But that convenience comes with a trade-off in flexibility and power. The platform's limited actions and siloed knowledge make it a poor fit for any team that needs to build automations that connect to the outside world. If your workflows depend on other tools, custom data, or knowledge stored outside of Atlassian, you're going to hit a wall pretty fast.

For businesses that are serious about using automation to improve how they work, a more specialized platform isn't just a better choice; it's a necessary one.

Build automations that actually work for your team

If you feel like the limitations of built-in tools are holding you back, it might be time to see what a dedicated AI platform can do.

You can connect your Jira Service Management and all your other knowledge sources to eesel AI in minutes. Start building the automations you actually need, without running into frustrating limits. Give the free trial a go today.