Daniele Montesi & Emanuele Viglianisi
summary
Scalable platform Zero2Hero runs on Azure and Databricks ingesting ~0.5 TB/day through Event Hubs into bronze/silver/gold Delta tables with ~200 ETL pipelines and ~2 minute data velocity. Pipeline extracts 500+ user features and computes labels, exposes two Databricks notebooks (feature impact and user-journey impact), and reports examples like a ~38 percentage-point uplift for a high-kilometre cohort and combined journeys summing to roughly a 50 percentage-point increase; CRM experiments confirm causation for at least one journey.
AI-written summary · about
chapters (12)
- 0:00Scalable data platform intro
- 1:41Runtastic and its data challenges
- 3:29Company strategy and vision
- 6:05Data sources and architecture
- 9:23Zero to Hero project overview
- 12:34Engagement metrics and segmentation
- 17:52Zero to Hero framework analysis
- 21:24Implementation and pipelines
- 27:05Solution value and CRM insights
- 32:28Correlation vs causation
- 34:36Conclusions and impact
- 35:19Q&A
related talks
55mSocial Listening for Richer Data, Unstructured Analytics
18mLessons from Analysing all of Europe's Data
33mData: The Land DevOps Forgot
27mHow to be data-driven when you aren’t Netflix (or even if you are)
14mCross-Cloud Campaign Analytics Meets JavaScript
36mData pipelines engineering made simple with Scala
29mData Acquisition: Overcoming Challenges
57mBillions of records: the real success story using Microsoft Power BI Embedded