Episode Summary:

Data Scientist Elliott Whitling joins us to discuss how marketing attribution vendors probably work behind the scenes. He outlines why you should ask the vendor how the vendor's algorithm learns and why you need to ask about identifying false positives. Then we look at what big data means for your anonymity. Finally, we dive into the differences between data scientists, data engineers and data analysts. 

Marketing Attribution Vendor Buying Guide:

  • How does the algorithm learn? Is the vendor using supervised or unsupervised learning?
  • How does the vendor identify false positives?
  • What is the rate of false positives? Ask for the ROC curve.
  • Understand the data going in, can the vendor technically access that data and will it give them enough to make an accurate model?

Show Outline:

0:00 - Intro / Introductions

1:02 - The data science of marketing attribution vendors

17:36 - Differential Privacy: Protecting privacy in large datasets.

28:40 - Data Scientist vs Data Engineer

34:08 - What is a large data set?

36:05 - Closing



Elliott's Blog

Differential Privacy

Apple and Differential Privacy 

Get Social With Us:

Still Unsponsored Facebook Page

Still Unsponsored on Instagram

Chris on Twitter

Chris on Instagram

Zach on Twitter

Zach on Instagram

Podbean App

Play this podcast on Podbean App