skills.rest
mattwg
mattwgCOMMUNITY
Explore all skills by this author →
GitHub

consumer-ds-metrics-lookup

Computes canonical B2C retention and acquisition metrics from Databricks SQL tables.

1|1|Updated Jul 31, 2026
Search Tags:#sql#retention#ltv#churn#databricks#subscription analytics#b2c metrics
One-click install
npx skills add https://github.com/mattwg/eureka-consumer-ds --skill consumer-ds-metrics-lookup-mattwg
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: consumer-ds-metrics-lookup
Source: https://github.com/mattwg/eureka-consumer-ds/tree/main/skills/data-science/consumer_and_degree_strategy_ds/consumer_ds/consumer-ds-metrics-lookup
Command: npx skills add https://github.com/mattwg/eureka-consumer-ds --skill consumer-ds-metrics-lookup-mattwg

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Consumer subscription metrics like cash, NPLs, NRLs, and M1/M2 retention are easy to compute inconsistently across analysts. This Skill returns each metric computed the canonical way the Consumer Strategy DS pod's dashboards do, using verified SQL against the correct Databricks tables. ## Core Features & Use Cases - Canonical Metric Definitions: Look up the exact formula, source table, grain, filters, and caveats for cash, NPL, NRL, new visits, total payers, retention rates, conversion, and LTV. - Verified Query Patterns: Run pre-validated SQL for step retention, 12-month survival curves, promo vs upsell vs full-price splits, channel cuts, 14-day registration-to-paid conversion, and completion-driven cancel vs churn. - Guardrails & Routing: Enforces B2C-only scope, aggregate-only output, cohort-size reliability flags, and declines out-of-scope questions (churn reasons, CAC, forecasts) with routing to the owning team. - Use Case: Ask "What are M1, M2 and M2+ retention for C Plus monthly, and where is the biggest drop-off?" and get the renewal percentages computed from base_data_for_ret_cancel with the standard eligibility filters applied. ## Quick Start Ask the agent for a consumer metric such as "Show M1 through M5 step retention for C Plus monthly for the January 2025 cohort by region, year over year."

Frequently Asked Questions about consumer-ds-metrics-lookup

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I calculate M1 and M2 subscription retention rates in SQL?▼

Compute retention as COUNT(next_txn_stamp_sub_level) divided by COUNT(*) grouped by payment_order on base_data_for_ret_cancel, filtered to B2C non-refunded BUY transactions with a 7-day eligibility lag. M2+ uses payment_order >= 2 without grouping.

How to measure new paying learners by marketing channel?▼

Sum npls from tof_consolidated_tracking_table filtered to user_segment='B2C', excluding financial-aid users and invalid traffic, grouped by referrer_cons_l1_mktg_chnl. Report only through the last complete month.

Why do new visits show zero when filtering by user segment?▼

In the TOF table, new_visits and visits are populated only on all-audience rows where user_segment IS NULL. Filtering or grouping visits by user_segment collapses them to zero; use segment filters only for NRL and NPL.

Can this skill explain why a retention metric dropped?▼

No, it returns the number computed the canonical way, not the cause. For root-cause analysis of metric movement, use the consumer-ds-metric-rca skill, which assembles on this one.

What are the limitations of this consumer metrics skill?▼

It covers B2C consumer metrics only and declines churn-reason text, data freshness issues, billing mechanics, PII lookups, B2B and degree metrics, engagement metrics, marketing spend, forecasts, and causal experiment readouts, routing those to the Consumer Strategy DS team.

Related Skills

Explore More Data & Analytics Skills→
SaberAlouiSaberAloui

database

Create and manage Replit PostgreSQL databases and execute SQL queries with safety checks.

Community
Intermediate
wachawowachawo

drilldown-analyzer

Decomposes aggregate metric data into dimensional segments to identify root causes.

Community
Intermediate
nmoralescybernmoralescyber

variance-analysis

Analyzes metric variance and outputs actionable explanations with driver breakdowns.

Community
Advanced
databrickslabsdatabrickslabs
38

databricks-metric-views

Create governed KPI metric views in Databricks Unity Catalog using YAML.

Official
Intermediate