Product Analytics
guides
Data Quality Monitoring: Schema Drift, Null Rates, Freshness SLAs, and Anomaly Detection for Analytics
Bad data causes bad decisions faster than good models fix them. This guide covers Great Expectations-style contracts, dbt tests, freshness SLAs on Airflow/Dagster, null and uniqueness monitors, schema evolution in Avro/Protobuf, volume anomaly detection (STL, EWMA), and how Airbnb-class analytics engineering triages P0 pipeline breaks vs metric definition drift.
User Segmentation & Behavioral Analytics: RFM, Clustering, Personas, and Production Guardrails
Segmentation powers targeting, pricing, and product prioritization — but k-means on raw features without scaling, leakage from future data, and unstable personas kill trust. This guide covers RFM, hierarchical clustering vs k-means, behavioral sequence features, evaluation metrics (silhouette with caveats), and how LinkedIn-scale teams ship segments with drift monitoring.
Product Analytics for Interviews: Metric Design, Root Cause Analysis, and Scenario Frameworks
The complete framework for product analytics interview questions — DAU drops, metric trade-offs, experimentation critique, and business case analysis. Covers the metric hierarchy (north star / guardrails / diagnostics), the 5-step root cause investigation process, common scenario traps, and how to structure your answer in under 3 minutes.
SQL for Data & ML Interviews: JOINs, Window Functions, and Query Optimization
Everything you need to solve SQL interview problems at data scientist, ML engineer, and data analyst roles. Covers JOINs and aggregations, window functions (ROW_NUMBER, LAG/LEAD, running totals), CTEs, NULL traps, and the query optimization patterns that separate strong from weak SQL answers.
Cohort & Retention Analysis: D1/D7/D30 Curves, Churn Interpretation, and Retention SQL
Retention is the backbone metric for subscription, consumer, and marketplace products — yet most candidates confuse rolling retention with fixed cohort retention, mis-handle right-censoring, and cannot write the window-function SQL. This guide covers cohort tables, classic vs unbounded retention, N-day windows, and how Meta-style teams diagnose false churn from product changes vs definition drift.
Funnel Analysis: Conversion Optimization, Drop-off Attribution, and Funnel SQL
Funnel analysis measures how users progress through a sequence of steps — signup → activation → first purchase → retention. It is the most frequently tested analytics framework in product data science interviews at Meta, Airbnb, Stripe, and Lyft. This guide covers funnel construction, drop-off rate calculation, multi-touch attribution, and the SQL patterns for session-level and user-level funnels.
Attribution Modeling: Last-Touch, Multi-Touch, Shapley, MMM, and Incrementality
Marketing and growth interviews ask how you credit conversions to channels — last-click is biased, multi-touch rules are arbitrary, and MMM is slow. This guide covers Markov attribution, Shapley-Owen values, geo holdouts for ground-truth incrementality, and how Uber-style teams pair fast digital attribution with slower marketing mix models.