We just wrapped a Speed up ticket triage with queues live learning session! We covered how to build and organize queues in Jira Service Management so agents can quickly find and prioritize the right work. We also looked at how AI features help at every stage, from before a request reaches the queue to while agents are working it.
Here’s a recap of the key takeaways, plus resources to help you put them into practice. Bookmark this page, try a few changes in your service space, and join the discussion to share what worked for your team!
────────────────────────────────────────
Step 1: Understand the queue’s role in triage
A queue is a filtered view of work items that agents use to triage requests. Think of it as a saved, always-updating list that shows only the requests matching specific criteria. Most teams need both shared queues (visible to all agents, for team-wide triage) and agent-specific queues (filtered to one agent’s workload, so individuals can focus).
How can AI help? A Rovo self-service agent can deflect requests before they ever reach your queues. It answers customer questions using your connected knowledge sources. If it can’t resolve the issue, it creates a ticket with the information it has already collected. That means your queues only hold the requests that truly need human help.
Tips for starting out
- Ask your team how they decide what to work on first. Your queues should reflect how your agents actually triage.
- Use a mix of shared and agent-specific queues. Shared queues give everyone visibility, and personal queues help each agent stay focused.
- Know that features shape your queues. Spaces with features like incident or change management turned on come with extra category-specific queues.
- Turn on a Rovo self-service agent for repetitive questions (like policy or how-to questions) to keep them out of the queue entirely.
Step 2: Build queues
Each queue uses conditions and a sort order: Think of conditions as the filter and sort order as the ranking.
- Conditions are JQL-based filters that check every request in real time. When a request’s fields match, it appears in the queue. When fields change, whether manually, through automation, or through a workflow transition, it moves automatically.
- Sort order decides what agents see at the top of the queue.
How can AI help? First, space admins can use Rovo Chat to list, create, and update queues using natural language, without needing to write any JQL.
Second, AI triage helps agents sort incoming requests faster. Requests from email or chat often arrive without a request type, or with a generic one. Agents can bulk-select those requests and have AI suggest the right request type based on each ticket’s summary and description. Once agents review and confirm, the tickets move to the right queue automatically. Because many queue conditions are based on request type, good queue design directly powers AI triage.
Tips for starting out
- Keep the default “All open” queue as a catch-all, and design your conditions so nothing slips through the cracks.
- Sort by what matters most. For time-sensitive work, try sorting by SLA time remaining, or by priority if SLAs aren’t set up yet.
- Start small and use your team’s language. Begin with a few queues that have clear names, and add more based on agent feedback.
- Keep conditions focused. Queues that return very large numbers of work items can affect performance, so limit them to recent work where you can.
- Be specific when using Rovo Chat. Open Rovo Chat and enter /manage-queue. Always include your space key or name, describe which tickets should appear, and ask for any extra columns you want. Then review the JQL that Rovo generates before you confirm.
Step 3: Organize queues
As a space grows, so does its list of queues, and agents shouldn’t have to scan a long list every time they open the sidebar. Space admins can use priority groups to organize queues into tiers. The top group is always visible, and the rest are collapsed below it. This is how you show agents what to focus on first. You can group queues by topic (for example, all contract-related queues) or by role or team (for example, queues for a group of specialists).
How can AI help? Once a request is in the queue, AI features can assist agents while they work it. With Suggestions turned on in Space settings, agents see a brief AI summary of the request and recommended next steps based on your knowledge base and similar tickets. With customer sentiment analysis on, Rovo shows whether the customer seems positive, neutral, or negative, so agents can respond with the right tone.
Tips for starting out
- Put your highest-volume or most urgent queues in the top priority group. Whatever is there is what agents see first. Everything else is one click away without competing for attention.
- Keep priority groups lean. Aim for around 5–10 queues per group. If a group has more than 10, it’s probably doing too much. The goal is less scanning, not just tidier clutter.
- Reorder your queues intentionally. The queues at the top of the list in Space settings are what agents see first.
- Use Rovo Chat for priority groups too. The same /manage-queue skill can list, create, and rename priority groups, and add queues to them.
- Try a sentiment-based queue. Sentiment can be used as a queue filter, for example to show requests where the customer seems negative.
Try it: Pick one queue your team uses every day and review its conditions, sort order, and priority group using the tips above. Then create a test request that should match its conditions and check that it appears where you expect.
NOTE: You must be a space admin of a company-managed Jira Service Management space to create and modify queues, and Rovo must be enabled on your site to use AI features. Anyone can use Rovo Chat to list queues and priority groups, but making changes requires space admin permissions.
👉️ Check out these resources when you’re ready to get started!
────────────────────────────────────────
Try something new this week
Pick one queue your team relies on and give it a quick checkup:
- Conditions: Does it show exactly the requests it should? Would anything slip through without the “All open” catch-all?
- Sort order: Is the most time-sensitive work at the top?
- Priority groups: Is it in the right group, and does your top group show what agents need to see first?
- AI features: Could a Rovo self-service agent keep some of these requests out of the queue? Could AI triage help route them here faster?
Then try asking Rovo Chat to list the queues in your space and suggest one change.
🌟 Share your experience below: Which queue did you check first, and what did you change? Which Jira Service Management AI feature are you keen to try out?
Keep learning
