Compare Rovo Chat reasoning modes

Rovo Chat can apply different levels of thinking to different types of asks:

The table below details the comparative qualities of each mode, when to use them, and how much time is typically required.

Rovo will activate think deeper mode automatically if it detects a multi-layered or open-ended prompt. You can also select it in your prompt if you know you want an analytical boost.

Deep research must always be explicitly selected as part of your prompt.

 

Basic Chat

Think deeper

Deep research

Best for…

quick answers and simple requests

  • summarizing

  • rewriting

  • explaining

complex questions requiring more structured output

  • measured reasoning

  • decision-making

  • multi-step analysis

heavyweight, evidence-backed, research

  • in-depth reporting

  • literature reviews

  • large cross-tool investigations

Qualities

  • Fastest response time

  • More structured, nuanced answers

  • Shows step-by-step reasoning

  • Medium response time

  • Leverages Teamwork Graph to cross-reference data across your organization

  • Shows step-by-step reasoning

  • Provides citations and traceability

  • Longest response time

Time on task

A few seconds

up to 2 minutes

up to 15 minutes

Examples

  • Summarize this page

  • Rewrite in a friendlier tone

  • Explain this error message

  • Which of these 3 roadmap options should we prioritize and why?

  • Compare approach A vs B and call out the tradeoffs

  • Identify 2 to 3 themes we should focus on for Q4, based on our Q3 goals and recent NPS feedback

  • Produce a competitive analysis of A vs B with pros/cons, pricing, and governance considerations

  • Analyze our last 6 months of incidents and produce a report on root‑cause patterns, MTTR trends, and recommendations

  • Draft a PRD or strategy doc based on our existing Jira epics, Confluence RFCs, and support tickets — with links to all sources

 

Still need help?

The Atlassian Community is here for you.