Data-driven Decision Making in Ad-Tech
summary
MIQ describes an end-to-end programmatic stack that connects 20 TB daily from 40+ data providers to build omni‑channel audiences, separate employees from customers with DBSCAN clustering, and swap underperforming strategies via dynamic feature selection feeding a recommendation engine. Bidding and budget use reinforcement learning to optimize DSP buys in real time. Claims include 40+ billion daily ad impressions traded, 500+ clients served, and a reported 54% incremental lift in campaign efficiency from their measurement‑driven optimization.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:45About MIQ and programmatic ads
- 4:14What is programmatic advertising
- 7:47MIQ’s data science solutions
- 8:52Omni-channel advertising
- 10:59ML use case on footfall data
- 12:22Campaign optimization strategies
- 14:07Measuring campaign effectiveness
- 16:36Business impact and results
- 16:57Summary and closing
related talks
36mML: Predictive analytics in Ad Engineering
38mContinuous Marketing Effectiveness Measurement
30mMMM and MTA: The Future of AI-Driven Marketing
46mFrom correlation to causation: measuring ad effectiveness at scale
15mThe Secret Life of Data - Part 2
47mIncrementality Bidding and Attribution
11mData science: Next-gen performance analytics
53mReal-Time Big Data Analytics with R