
Performance Data Interpretation & Insight Synthesis
Transform raw metrics into actionable performance insights
What You Can Do
You can upload or paste performance data in any format (CSV, JSON, spreadsheet exports) and Claude will analyze patterns, identify correlations, detect anomalies, and synthesize insights into clear, strategic recommendations. The skill interprets complex metrics across marketing, sales, product, and operations domains, explaining what's driving performance changes and suggesting evidence-based next steps.
Features
Identifies unusual spikes, dips, or patterns in your data and explains the statistical significance and potential causes
Discovers relationships between different metrics (e.g., campaign spend vs. conversion rate) and evaluates whether relationships are causal
Projects future performance based on historical patterns and seasonal trends, with confidence intervals
Contextualizes your metrics against industry standards, competitors, or previous time periods to clarify performance standing
Synthesizes data across multiple dimensions to pinpoint underlying factors driving performance outcomes
Creates clear, jargon-free summaries for non-technical stakeholders with key findings and recommended actions
Analyzes performance across customer cohorts, channels, geographies, or product lines to isolate strengths and weaknesses
Example Output
Input: CSV with monthly revenue, customer acquisition cost, churn rate, and marketing spend for the past 18 months
Output:
Key Finding: Revenue is flat despite 40% increase in marketing spend. Root cause: customer acquisition quality declined. Average customer lifetime value dropped 35% — acquired customers from new channels churn 3x faster.
Supporting Evidence: CAC rose from $45 to $82, but first-month retention fell from 87% to 64%. Lower-quality cohorts correlate with discount-driven campaigns in months 6-10.
Recommendation: Reallocate budget from discount campaigns to brand-awareness channels. Test 2-3 messaging variations to recover customer quality. Expect 2-month lag before revenue impact.
Confidence: High (strong correlation across three independent metrics)
Another example: You upload product analytics showing time-on-feature dropped 45% for your flagship feature last week. Claude identifies that this correlates with a UI redesign on Tuesday and flags that similar metrics declined for 2 other features on the same day, suggesting a systemic UX issue rather than product-market fit deterioration.
What's Included
- Multi-format data parser: Automatically detects and interprets CSV, JSON, spreadsheet exports, plain text, and screenshot data without manual reformatting
- Statistical interpretation framework: Applies standard statistical methods (correlation, trend analysis, anomaly detection) explained in plain English
- Domain-specific glossaries: Reference materials for marketing, sales, product, and operations metrics so Claude uses correct terminology and benchmarks
- Insight synthesis templates: Structured approaches to translating data findings into strategic recommendations and action items
- Visualization interpretation guide: Tips for extracting insights from charts, graphs, and dashboards by describing them to Claude
Who It's For
- Data analysts and BI professionals
- Product managers and directors
- Marketing leaders and campaign strategists
- Sales operations and revenue leaders
- Executive stakeholders needing clear metric interpretation
Best For
- Monthly or quarterly performance reviews
- Explaining unexpected metric changes to leadership
- Identifying root causes of performance dips
- Forecasting future performance trends
- Comparing performance across segments or time periods







