Coub: how we built an analytical platform

on YouTube ↗36m

" summary

Coub built its own analytics platform to process and analyze about 1 billion events per month from players, an iOS app, and the website. The talk explains why third-party services were too costly and limiting at that scale and describes the in-house solution they designed and have been using for over a year to store and analyze those events.

AI-written summary · about

" chapters (11)

1. 0:00Introduction
2. 3:34Why analytics
3. 6:07Writing our own
4. 9:59First prototype
5. 11:07New platform requirements
6. 13:53Choosing analysis tools
7. 19:16MongoDB data structure
8. 22:26Current system architecture
9. 24:58Product analytics use
10. 26:45Future plans
11. 27:29Q&A

" related talks

• 04mInterview with Gyula FóraGyula Fora · Flink Forward Berlin 2016
• 27mBig Data in Mobile GamesAndre Cohen · Games Gathering 2020
• 10mon their new analytics platformPatrick Lightbody · Velocity Santa Clara 2015
• 55mCloud Analytics: from 0 to 100,000 events per second in 50'Didier Girard · DevFest Nantes 2016
• 30mHarnessing Power: Inside Sofascore’s Data-Driven InfrastructureKarlo Knezevic · DSC Europe 24
• 18m1,200 analytics requests a day: Clay and Vercel's Bet on AI-Native AnalyticsAbhi Sivasailam, Josh Hanson · Vercel Ship 26 New York City
• 19mA History of Automatic AggregationsRaphaël Bizos · PromCon EU 2025
• 27mAnkur Sharma: Head of Analytics and User Growth at Instamojo speaks at Cypher 2017Ankur Sharma · Cypher 2017