What if you could update your CRM record in Jira after a customer meeting just by speaking? Or ask AI which deals need your attention today, instead of manually reviewing your entire sales pipeline?
And what if your AI assistant could go even further - connecting sales information with technical tasks in Jira to help you understand what's really happening with a customer?
These are some of the scenarios we've been exploring with Atlassian Rovo and Mria CRM Assistant.
Today, I'd like to share three practical use cases showing how AI agents can make everyday CRM work in Jira easier, from meeting follow-ups to more informed decisions across sales and delivery teams.
But first, a little context.
What Are Atlassian Rovo Agents?
If you're already using Jira or Confluence, you've probably heard about Atlassian Rovo. Rovo brings AI-powered search, chat, and agents into the Atlassian ecosystem, helping teams find information, understand context, and take action without constantly switching between tools.
Rovo Agents take this a step further.
Think of them as specialized AI teammates. Instead of simply answering general questions, they can be equipped with specific knowledge and tools to perform tasks within your workflows. For example, an agent might help you organize Jira work items, summarize project progress, prepare documentation, or work with information from connected applications.
With Rovo Studio, teams can also create agents tailored to their own processes.
This opens up interesting possibilities for Marketplace partners, too. By extending Rovo with app-specific capabilities, we can make AI useful for workflows that go beyond standard Jira or Confluence functionality.
That's exactly what we've been working on with Mria CRM Assistant, a Rovo agent integrated with our Mria CRM. Let's look at some practical examples.
Use Case 1: From Meeting Notes to Follow-Up Tasks - All with One Voice Request
We've all been there…. 🙂 You finish a customer meeting, have several important points to capture, and need to update your CRM before moving on to the next call. Usually, that means finding the right deal, opening its details, writing notes, and creating follow-up tasks with deadlines. But what if you could simply tell your CRM what happened?
Example scenario
Imagine you've just finished a meeting with a potential customer. Instead of manually navigating through CRM records, you open Rovo Chat, select Mria CRM Assistant, and simply dictate:
"Hey Rovo, please find the Demo Deal and add a note summarizing our meeting from yesterday, November 2.
The customer has 1,000 users and is interested in an annual license. The main blocker is customer reporting, which we plan to deliver next week.
Once you've added the note, please create two follow-up tasks and assign them to me:
- Send the customer a delivery notification by the end of this month.
- Send an email reminder about the free trial next Friday."
With a single voice request, Mria CRM Assistant can:
- Find the Demo Deal in Mria CRM.
- Add a note capturing the key meeting details, including the customer's requirements and the outstanding blocker.
- Create two follow-up tasks with the requested deadlines and assign them to you.
- Ask for your confirmation before saving the changes.
Why is this useful?
One voice request replaces several manual CRM actions: searching for a deal, documenting a meeting, creating follow-up tasks, setting deadlines, and assigning ownership.
And the best part? You can do all of this in natural language, without opening multiple CRM screens or filling in forms.
It's a simple example of how AI can help keep CRM data up to date while letting you focus on actual customer conversations.
Note: Voice input uses your device's speech-to-text functionality, while Rovo processes the resulting text.
Use Case 2: Find Leads and Deals That Need Attention and Decide What to Do Next
One of the challenges in sales isn't necessarily managing a large number of leads. It's knowing which ones deserve your attention right now. A CRM can show you lists, statuses, dates, and pipeline stages. But you still need to review that information and decide what to do. This is where an AI assistant can help.
Example scenario
It's Monday morning, and you want to plan your sales activities for the week. Instead of opening multiple CRM views and manually filtering records, you ask:
"Review my open leads and deals. Find those that haven't been updated in the last 14 days, especially high-value opportunities or deals approaching their expected closing dates. Suggest which ones I should follow up with first and explain why."
Mria CRM Assistant can search records using available CRM data, including deal stages, amounts, owners, dates, and other fields.
Based on the results, you can ask it to help prioritize your next steps.
You can then continue:
"Suggest a short, friendly follow-up message for each of these deals, based on the available context."
Or take it a step further:
"Create follow-up tasks for these three deals and assign them to me."
The assistant prepares the changes for your confirmation.
Why is this useful?
This moves CRM from simply storing information to helping you make decisions. Instead of asking, "What deals do I have?", you can ask, "What should I focus on today, and why?"
Of course, AI recommendations still need human judgment. A deal that hasn't been updated recently isn't necessarily at risk. But having an assistant identify potential gaps and suggest actions can make pipeline reviews much more efficient.
Use Case 3: Connect Sales and Delivery Context and Understand the Full Customer Picture
This is probably my favorite use case, and the one that best demonstrates why having CRM inside the Atlassian ecosystem matters.
In many B2B companies, customer relationships extend far beyond the sales pipeline. Sales teams manage opportunities and customer conversations. Delivery teams work on implementations, technical requirements, onboarding, and support requests in Jira. And these activities are often directly connected. A deal might depend on a technical integration being completed, a proof of concept being validated, or an outstanding customer requirement being resolved.
But how often does the person managing the customer relationship have a complete picture of what's happening on the delivery side?
Example scenario: Preparing for a customer meeting
Imagine you're an account manager preparing for an important call with a customer. The customer has an active deal in Mria CRM, several technical tasks linked to it in Jira, and ongoing discussions about implementation requirements.
You want to understand whether everything is progressing as expected without opening every related record and work item individually.
So you ask Mria CRM Assistant:
"Review the Acme CRM Migration deal and its linked Jira work items. Summarize the current deal status, identify any outstanding technical tasks, and highlight anything that might affect our ability to close this opportunity."
This is where combining CRM and Jira context becomes particularly valuable.
The commercial perspective from CRM:
- Current deal stage (Won), deal value, and closing date.
- Deal ownership and pipeline information.
- Recent and upcoming customer activities.
- Customer notes and interaction history.
The delivery perspective from Jira:
- 12 associated Jira work items and their statuses.
- Tasks currently in progress, including CRM-117 (Activities panel improvement).
- Outstanding tasks, including MCD-1172 (Customize Jira work item panel) and MCD-1174 (Auto-save email drafts).
- Technical work that may require follow-up or affect customer commitments.
With the appropriate Jira context available to Rovo, the assistant can help connect these perspectives and provide a more complete picture of the customer situation.
Going beyond a simple status summary
The real value isn't just getting a summary of multiple records. It's being able to ask questions that require understanding the relationship between commercial commitments and technical execution.
For example:
"Are there any unresolved technical issues that could delay this deal?"
Or:
"Which customer requirements are still pending, and who is responsible for them?"
Or even:
"Based on the current progress, what should I clarify with the customer during our next meeting?"
Consider a situation where a deal is progressing well commercially, but a critical integration task in Jira is still blocked. Without visibility into delivery, the account manager might continue discussing the expected closing date without realizing there's a potential risk.
On the other hand, the technical team might have already resolved an important customer concern, creating an opportunity for sales to move the deal forward.
Why is this useful?
Because the real customer context rarely lives in a single record or tool. It's spread across sales conversations, support requests, implementation tasks, technical discussions, and project updates. And this is exactly the gap we're trying to address with Mria CRM.
By bringing CRM into Jira and making that information accessible through Rovo, we're exploring a more connected way to manage customer relationships. Instead of treating sales and delivery as separate processes, teams can work with a shared customer context.
For me, that's a much more interesting application of AI than simply generating summaries or drafting emails. It's about helping people understand what's happening across teams and make better decisions based on that information.
Rovo AI + CRM: From Finding Information to Taking Action
Looking at these three examples, I think there's an interesting progression in how AI can support CRM workflows in Jira.
1. Capture and update information.
Turn meeting summaries into structured CRM notes and follow-up activities without manual data entry.
2. Analyze and prioritize.
Identify opportunities that need attention, review available customer context, and suggest next steps.
3. Connect information across teams.
Bring together commercial and delivery perspectives to understand customer situations more completely.
And importantly, AI doesn't have to stop at providing recommendations.
With Mria CRM Assistant, you can also perform supported CRM actions, such as creating records, updating information, adding notes, and managing activities. Data-changing actions require user confirmation before execution, and the assistant respects the permissions of the person using it.
I believe this balance is important: AI should help people work faster while keeping them in control of business-critical information.
How to Try Mria CRM Assistant
If you're using Mria CRM for Jira and Rovo is enabled on your Atlassian site, the assistant is already available - no separate agent installation or configuration is needed. Simply open Rovo Chat -> find Mria CRM Assistant in the Agents list -> and start asking questions in natural language.
You can learn more about its capabilities, supported actions, and limitations in our Mria CRM in Rovo documentation.
Final Thoughts: AI Is More Valuable When It Understands Your Workflow
There are already plenty of AI tools that can summarize text, generate responses, or help automate repetitive tasks. But I believe the real value comes when AI understands the context of your work and can help you act on it.
For teams using Atlassian, that means connecting customer information, sales activities, technical delivery, and collaboration in a more natural way. That's also why we see Rovo as much more than an AI chatbot. It's an opportunity to make the information and workflows already living in Atlassian more accessible and actionable.
We're continuing to explore what's possible with Rovo AI and Mria CRM, particularly around reducing administrative work and improving collaboration between sales, support, and delivery teams.
I'd love to hear how others in the Atlassian Community are using Rovo Agents!
Are you mainly using them to find and summarize information, or have you already started building workflows where agents take action across your tools? And if you could connect AI to your CRM and Jira workflows, what would be the first task you'd delegate to it?
