Generating targeted AI revenue at mass scales
summary
Air Miles builds an in‑house personalization system called Smokey platform that uses an ensemble of collaborative filtering, deep learning, reinforcement learning and a contextual bandit ensemble tied to a cloud data lake with automated ETL and marketing APIs. It deploys about 40 million offers/month, reports average campaign ROI gains of 7–10% with some campaigns improving marketing ROI by ~240% and email opens up ~43%, after years and millions invested and major organizational change.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 4:01Channels of communication
- 5:15Optimizing for ROI
- 9:20Targeting
- 10:50Key benefits
- 11:58Marketing campaign ROI
- 12:17Email open rates
- 15:12Characterism
- 17:40Risk
- 18:25Leap of faith
- 19:52Talent
- 21:06Business Processes
- 22:16Questions
related talks
28mHow Air Miles drives end-to-end marketing solutions
30mFrom €1M License to In-House Success: How We Built a Real-Time Recommendation System and Saved Millions Doing It
30mAI for Customer Service and Marketing at Aeromexico
23mThe unreasonable effectiveness of data for better monetisation
19mMachine Learning is a new age of Personalized Game Monetization
24mKeynote: AI Trends in Performance Marketing
50mTake AI/ML from theory to practice with Intel technologies
29mAI and ML for Computational Advertising