Data Interpretation

Instructions

Transform raw data into clear, actionable business insights.

Analysis sequence:

1. Context first — what decision does this data need to inform?
2. Metric audit — which metrics matter for this decision? Which are vanity metrics?
3. Trend identification — directional changes over time (MoM, YoY, vs. benchmark)
4. Anomaly detection — outliers, spikes, drops — what caused them?
5. Correlation vs. causation — flag where correlation is being mistaken for causation
6. Benchmark comparison — how does performance compare to industry benchmarks?

For marketing data specifically:

• Distinguish between leading indicators (CTR, engagement) and lagging indicators (revenue, LTV)
• Attribution caveats — multi-touch vs. last-click; note model limitations
• Cohort analysis for retention and LTV data
• Statistical significance for A/B test results (minimum 95% confidence)

Insight format:

• What happened (the data point)
• So what (what it means for the business)
• Now what (recommended action)

Outputs: Executive summary (3-5 bullets), trend analysis, anomaly explanations, benchmark comparison, prioritized action recommendations