mattwg
Community@mattwg
Consumer subscription analytics skills for canonical retention metrics, promo campaign recaps, and pricing-test impact analysis on Databricks data.
Agent Skills by mattwg
Showing 3 vetted skills indexed across 1 GitHub repositories.
mattwgconsumer-ds-metrics-lookup
Computes canonical B2C retention and acquisition metrics from Databricks SQL tables.
mattwgpricing-test-impact-on-promo
Analyzes how an A/B pricing test confounds a concurrent promo using SQL and matplotlib trendlines.
mattwgpromo-campaign-recap
Computes standardized promo campaign performance cuts from Databricks transaction and Amplitude data.
Frequently Asked Questions About mattwg
FAQPage SchemaWhat tasks can I accomplish with mattwg's consumer analytics skills?▼
You can compute canonical B2C metrics (cash, NPLs, NRLs, retention rates, LTV, conversion) with segmentation, generate standardized promo campaign recaps across regions and channels, and measure how a pricing A/B test interacts with a concurrent promo, including phase-level cash-per-user splits.
Who are these skills designed for?▼
They target consumer data science and strategy analysts working on subscription businesses — specifically teams measuring learner acquisition, retention, promo performance, and pricing experiments, as indicated by the consumer_ds pod ownership and Coursera-specific tables and campaign taxonomies.
What are the runtime requirements and dependencies?▼
The promo-campaign-recap skill requires the Coursera Data MCP (MintMCP) connected to a Databricks MCP server for live query execution against tables like completed_carts, payment_order, and tof_consolidated_tracking_table. The other skills rely on the same underlying consumer data warehouse schemas.
How does the pricing test skill handle promo confounds?▼
It isolates the EPIC Test/Control population using impression windows, splits Cash/Users/Cash-per-user by promo phase (Pre-promo, Early Bird, Post-Early-Bird), runs a 50:50 traffic-split sense check by channel and country, excludes urgency messaging, and calculates final USD impact.
When should I use promo-campaign-recap versus a single metric lookup?▼
Use promo-campaign-recap for multi-campaign comparisons or quarterly post-mortems with standard cuts by region, channel, and page. For a single ad-hoc metric pull, the manifest directs you to promo-metrics-lookup; for diagnosing why a metric moved, use promo-metric-rca.