Delivering Retail Analytics & Data Projects With Data Created For Purpose
summary
Claim: creating purpose-built event data with Snowplow and modeling it in dbt/SQL Runner cuts weeks of cleaning and flips work from ~80% wrangling to ~80% value delivery. Implementation: instrumented JSON event schemas with versioning, custom contexts, sessionization and user-stitching, landed into Snowflake (or BigQuery/Redshift) and transformed by dbt packages and SQL Runner into pageview/session/user tables. Result: schema evolution and billions of events ingest without breaking models, restoring metric trust.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:05Data science project failures
- 4:57Behavioral data explained
- 7:02Tracking user interactions
- 12:57Sessionization challenges
- 15:47Data modeling benefits
- 20:20Business case for data
- 22:05Q&A
related talks
18mBehavioral Data Creation for AI - Delivering Value and Insights
26mPutting your data warehouse to work - Reverse ETL & Operational Analytics
01h 17mData Engineering in the AI Era: Building a Robust Foundation for AI Applications
39mBREAKING YOUR DATA TEAM OUT OF THE SERVICE TRAP
23mData Team as an Optimisation Problem
31mAdvancing security analytics with Snowflake and AWS at Snyk (PRT329)
14mA Brave New World: How New Lineage Technology Can Reduce Data Spend 40%
49mTransforming Analytical Data with Event-Driven Architecture