HOW AI SETS THE AUDIENCE HAYSTACK ON FIRE Lance Schafer General Manager Product & Technology
Types of AI will benefit programmatic advertising PART I: Desired outcomes of machine learning (and risks) PART II: Models used in automotive programmatic AI PART III: Example of AI vs human-deployed programmatic PART IV: Takeaways PART V:
THE AI LANDSCAPE - WHICH WILL DRIVE PROGRAMMATIC ADVERTISING? 4
WHAT IS THE DESIRED OUTCOME OF ML POWERED PROGRAMMATIC ADVERTISING? Automatically learn and improve from experience without explicit programming ● Incorporate more data for decisions ● Weather ● Consumer Confidence Indicators ● Ever increasing better results
CHALLENGES OF MACHINE LEARNING Not being able to Algorithmic risk communicate the WHY? ! Why did you buy those ads? Because the machine told us to!
MODELS USED IN AUTOMOTIVE PROGRAMMATIC AI Major learning - Take advantage of all the structure of automotive to combine wide models with deep learning 7
TAKEAWAYS CASE STUDY RESULTS ● ML Objective ○ Keep volume same ○ Lower cost per goal ● Other metrics measured ● Time on site ● Pages per session 8
OUTCOMES Costs Goals Time on site Pages per lowered increased increased session lowered by 17% by 28% by 9% by 16% 9
TAKEAWAYS TAKEAWAYS We should embrace We should not assume our wide & deep hybrid customers understand AI models and ML, and its limitations 1 0
1 1
Recommend
More recommend
