Summary
Challenge: A high volume of business-critical data quality requests makes it difficult to maintain SLAs, especially during peak periods. This leads to slower resolution times and increased reliance on human intervention.
Solution: Four Rovo Agents collaborate to triage queries, identify issues, suggest fixes from Confluence, and populate ticket fields. A Data Quality Engineer reviews outputs, with everything built in Rovo Studio using natural language prompts.
Outcome: This improves efficiency and speeds up handling of critical requests. It reduces manual work, supports peak demand, and frees up human agents for higher-value tasks.
Background
In our work with longstanding client Convex, an international speciality insurance firm, we’re continually exploring how we can automate and enhance internal processes for its business teams. Convex has embraced Rovo across its Atlassian toolset – particularly in Jira Service Management (JSM).
We have created a range of custom Rovo Agents to unlock greater time efficiencies, tighter quality control, and happier users across the organisation.
In this case study, we’re focusing on a custom Atlassian Rovo Agent to improve data quality request handling.
Challenge
Data quality is essential for an insurance firm, and the Convex Data Quality team handles a significant volume of queries for all systems across the business.
These each need to be dealt with as swiftly and effectively as possible – especially at times of peak system usage. Convex also wanted to free up human agent time to focus on the highest priority requests, and to automate less complex ones.
Solution
The Data Quality team already used Atlassian Assist to create service desk ticket from requests that are raised in a Slack channel. We built a group of four Rovo Agents which work together to strengthen the next part of the process: Triage.
How do the Data Quality request triage Rovo Agents work?
The solution comprises one lead Rovo Agent, and three supporting ones:
The first Agent reviews the request, searches for related issues and identifies the issue.
By parsing knowledge base articles and content, held in a specific Confluence space, the second Agent suggests a fix for the Data Quality Engineer.
This Agent takes care of the ticket custom fields, of which there can be up to 12. These fields might include ‘Severity level’, ‘Systems impacted’, and more…
The final Agent handles the formatting of those custom fields.
With all four custom Rovo Agent combined, our client’s Data Quality team now benefits from a robust, automated request triage process.
Before a ticket is closed, a human agent will have the opportunity to edit and approve all fields. We proactively built this approval step into the workflow, as it was important to Convex that a human sign off all agentic input.
Benefits
This custom Rovo Agent now handles all simple Data Quality queries, automatically suggesting a fix and populating ticket fields.
This has freed up significant human time, so the Convex Data Quality Engineers can focus on the more complex, high priority cases.
