Triangulating data and experiments beats trusting one source

14 Nov 2025 · 0 min
0:07

Michael Kaminsky and a co-speaker argue that no single vendor can solve advertising measurement. Real insight requires combining multiple data sources and custom analysis, not trusting promises of a universal solution.

More from Michael

5:33
youtube.com

A/A Testing: How to Check If Your Experimentation Platform Actually Works

2mo ago
youtube.com

How to Actually Check If Your Media Mix Model Is Right

3mo ago
youtube.com

The Continuous Testing Loop That Separates Great Marketing Teams From Average Ones

3mo ago
youtube.com

Why "MMM Is Only a Report Card" Completely Misses the Point

4mo ago
youtube.com

The Difference Between Predicting What Happens and What You Make Happen

4mo ago
youtube.com

How Hierarchical Models Fix One of the Biggest Flaws in Media Mix Modeling

4mo ago
youtube.com

The Difference Between Attribution and Incrementality Explained

4mo ago
youtube.com

Statistically Significant Doesn't Mean Worth the Spend

4mo ago
youtube.com

Plato's Cave and the Lie Your ROAS Is Telling You

4mo ago
youtube.com

Why "Statistically Significant" Results Can Still Be Completely Wrong

4mo ago
youtube.com

How to Actually Test If Your Meta Ads Are Working

4mo ago
See all content