Insights help transform raw research data into clear, actionable recommendations your team can use to make decisions. In Marvin, you can analyze research, identify patterns, and turn findings into polished reports that stakeholders can easily consume.
The goal is not only to document research; it’s to drive decisions and action.
Follow the steps below to create your first insight:
Before you start: If you're starting your journey with Marvin, we recommend reading the Getting started with Marvin Projects guide before diving into insights.
Step 1: Add research data to a project
Start by uploading or connecting your raw research data inside a project workspace. At this stage, your primary focus should be collecting user evidence without interpretation.
You can add and centralize data including:
Transcripts and recordings: Interview transcripts, video, and audio clips.
Unstructured notes: General research notes, bullet points, and user highlights.
Quantitative data: Survey responses, behavioral logs, or usability data metrics.
Learn more:
Step 2: Review and analyze your data
Once your data is successfully uploaded, you can leverage Marvin’s suite of analysis tools to automatically or manually surface patterns and recurring themes.
Consider how you want to approach your synthesis:
Thematic Analysis: Automatically surface recurring themes and calculate mention frequency across your data.
Ask AI & Deep Research: Run quick questions across your files or execute advanced synthesis like persona and JTBD validation.
Q&A Tables & Affinity Maps: Visually cluster related notes or organize cross-participant survey responses side-by-side.
Learn more:
How do I analyze the data in my project?
Best practices for using Deep Research.
Affinity mapping in Marvin
Step 3: Turn findings into insights
Findings describe what users said or did (e.g., “Multiple participants struggled to locate the checkout button”). Insights take it a step further, explaining why those findings matter and mapping out the exact actions your product team should take next.
Use Marvin's core features to measure and connect your findings:
Quantify your data: Use auto-tagging, labels, and built-in charts to measure exactly how often themes recur across all interviews.
Ground your findings: Use direct citations to link your discoveries back to the exact timestamped transcript, note, or video clip.
Note: AI can suggest patterns and automated groupings, but researchers always maintain full control to review and validate findings before they are finalized.
Learn more:
Step 4: Build your Insight report
Now that your findings are validated and supported by data, it's time to build a polished, comprehensive report that your stakeholders can easily absorb.
To customize your report:
Click the light bulb icon to generate a new insight report or add to an existing one.
Edit text, organize your document sections, and add structural headers or context.
Drag and drop interactive video clips, charts, and visualizations directly into the layout so your audience can see the literal voice of the customer.
Learn more:
How to create Insights?
Step 5: Publish and share your report
Don’t let your research live in a vacuum. Once your report is complete, bring your team into the workflow and share the finished product with key decision-makers.
You can collaborate and distribute via:
Draft reviews: Share your in-progress insight reports with teammates to gather internal feedback.
Discover Research hub: Publish finalized reports to your shared company repository so anyone in your workspace can access them.
Integrations: Push automated updates directly to your team's Slack channels, or convert your published report into an AI-generated Slide Deck.
Pro tip: We find the most impactful research reports rely on interactive citations. Allowing your executive stakeholders to click a finding and instantly watch the 15-second customer video clip drastically speeds up executive alignment and product buy-in.
Learn more:
How do I collaborate with my colleagues?