Skip to main content
  1. Start broad, then narrow: use list tools first, then search with specific filters.
  2. Chain queries: get a study ID, then find its candidates, then retrieve their transcripts.
  3. Use full-text search: most search tools support searching within titles, descriptions, and content.
  4. Leverage pagination: for large result sets, use pagination to process data in batches.
  5. Combine filters: most search tools accept multiple filters for precise results.
  6. Reach for templates before building from scratch: search or list your saved templates first, then apply the one that fits instead of rebuilding a screener question by question.
  7. Let the AI chain writes: describe what you want end-to-end (e.g., “set up a study with a screener and invite these candidates”) and let your AI tool handle the individual tool calls.
  8. Search the whole repository first: ask a repository-wide question before narrowing to transcripts. Your team may already have highlighted the key moments or written an insight on the topic, and building on that beats re-deriving it from raw sessions.
  9. Reuse your team’s tags: ask your AI tool to look up existing tags before creating new ones, so agent-coded work stays consistent with how your team already organizes research. Tag names must be unique within a study, so a duplicate is rejected and the existing tag is suggested instead.
  10. Highlight by quote, not timestamp: when creating highlights, have your AI tool quote the passage word for word from the transcript. Transcripts show only minutes and seconds, so quotes are more reliable than time ranges.
  11. Check what your agent saved: everything written through the MCP is saved under your name, with visibility governed by existing access permissions. Open the highlight, reel, or insight in Great Question (results include app links) to confirm it says what you meant before sharing it.
Still need help? Contact us at support@greatquestion.co — median response time is 19 minutes during support hours.