Explain performance data with Performance Insights

Analyze engagement and content evidence from a connected dataset and emit structured performance findings plus a report.

Where: Flowgen → Add → Research → Performance Insights

Key ideas

• Evidence in, findings out: The node consumes a compatible dataset and produces structured insights without changing the underlying rows.
• Analyst model: Choose the analysis model from the same card/sidebar control; preserve the selected model when comparing runs.
• Interpretation, not attribution: The result can describe associations in the collected data but cannot prove that one creative choice caused the performance.
• Performance Insights card: PerformanceInsightsNode uses the generic shared-card path. Analyze a connected dataset for performance findings and a readable report.
• Properties sidebar: researchNodes.tsx
• Opened surfaces: Structured result preview

Steps

1. Connect a dataset that contains the metrics and content fields relevant to the question.
2. Choose the analyst model and run the node.
3. Review findings against individual rows, outliers, missing metrics, platform differences, and sample size.
4. Connect supported findings to a plan, brief, QA checklist, or Ask the Data follow-up.

Performance Insights controls and presets

• Add path: Flowgen → Add → Research & Insights → Performance Insights.
• Card: PerformanceInsightsNode.
• Properties: researchNodes.tsx.
• Controls: Dataset input; analyst model; structured insights; report.
• Opened surfaces: Structured result preview.

Tips

• Compare like with like: similar formats, platforms, periods, and audience conditions.
• Ask a narrower follow-up in Ask the Data when an aggregate finding hides exceptions.

Limitations and important notes

• The node reports patterns in the supplied dataset; it does not establish causality or correct source-data bias.

Related guides

• Choose a Flowgen research or analysis node
• Answer a focused question with Ask the Data
• Collect posts with Content Harvest