Superstar Session: Hypercasual Live Ops- Building a Data-driven Growth Machine
summary
Defines live ops as post-scale (about one month after scaling) and describes a data-first pipeline that runs hundreds of parallel tests across stores, countries and traffic sources, with results aimed in about seven to ten days. Uses Firebase + Remote Config feeding BigQuery and in-house Python models and a dashboard to score experiments automatically. Notes sample sizes of ~5k–30k per group, cumulative 5% lifts compounding to ~1.8x LTV, and heavy automation for rollout and QA.
AI-written summary · about
chapters (23)
- 0:00Intro
- 1:26Introduction
- 1:40Paper Plan
- 2:16Method
- 2:58Agenda
- 3:14Crazy Loves
- 3:35Recent Games Success
- 3:47What is Live Ops
- 5:50Positive Feedback
- 6:19Motivation
- 7:29Personal Confession
- 8:59Test Results
- 9:59Under the Hood
- 14:08The Secret Weapon
- 15:20Mini Goals Machine
- 16:08Recap
- 16:21Questions
- 21:33Testing system
- 22:23Prototype vs AB test
- 23:13Final result
- 24:03Crazy Labs
- 24:53Internal Discussion
- 26:10Challenges
related talks
49mThe Analytics of Growth Hacking
27mHow to be data-driven when you aren’t Netflix (or even if you are)
27mUnlocking Product Growth With Big Data, Data Science And Machine Learning
42mBuilding a Data Driven Organization
50mData-Driven Engineering
19mData & Analytics; Shaping the Future of Gaming and Maximizing ROI
16mSuperstar Session: Introducing Benchmarks+ Performance Insights for 1.2bn Monthly Players
41mEnhancing Event Analysis at Scale: Leveraging Tracking Data in Sports