- Home
- Skills
- Anthropics
- Knowledge Work Plugins
- Performance Analytics
performance-analytics_skill
- Python
- Official
7.4k
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill anthropics/knowledge-work-plugins --skill performance-analytics- SKILL.md14.5 KB
Overview
This skill analyzes marketing performance across channels and campaigns to surface key metrics, trends, and optimization recommendations. It provides channel-specific metric definitions, benchmarking guidance, reporting templates, and a repeatable process for trend analysis, attribution, and testing. Use it to turn raw data into prioritized actions that improve marketing ROI and pipeline efficiency.
How this skill works
The skill inspects channel data (email, social, paid, SEO, content) and maps metrics to funnel stages and business outcomes. It identifies directional trends, inflection points, seasonality, and anomalies, then applies simple forecasting and attribution logic to generate hypotheses. Finally, it produces prioritized optimization steps and reporting templates (weekly, monthly, quarterly) tailored to decision cadence.
When to use it
- Preparing weekly, monthly, or quarterly performance reports for stakeholders
- Reviewing campaign results to identify what's working and what needs improvement
- Analyzing channel health (email deliverability, social engagement, paid efficiency, SEO visibility)
- Building forecasts or scenario models for short-term planning
- Designing and prioritizing optimization tests across the funnel
Best practices
- Lead reports with metrics tied to business objectives and show trends over time
- Compare metrics to prior period, targets, and benchmarks for context
- Start attribution with last-touch if no model exists, then compare multi-touch perspectives
- Prioritize tests by impact and effort: high-impact/low-effort first
- Test one variable at a time, predefine success metrics, and run for a full business cycle
- Present forecasts as ranges and flag low-confidence predictions when data is limited
Example use cases
- Generate a weekly marketing snapshot for standups listing top metrics, wins, and priorities
- Audit an underperforming email program using deliverability, open, CTR, and conversion benchmarks
- Assess paid campaigns to calculate CPC, CPA, ROAS, and recommend budget reallocation
- Perform trend analysis on organic traffic to identify seasonal patterns and top entry pages
- Develop a prioritized A/B testing roadmap to improve funnel conversion rates
FAQ
Start with last-touch for simplicity and actionability, then compare with first-touch and position-based models to understand awareness vs. conversion drivers.
How many data points do I need for trend analysis?
Aim for at least 8-12 periods for meaningful directional trends; fewer points reduce confidence and should be labeled low confidence.