In this episode we discuss the fundamentals of microtargeting, voter data mining and predictive analytics. We talk about why this data is so valuable, how we generate it and how it can be used to give a campaign an edge on their opponent.

Segment #1 – The End Result of Predictive Modeling

  • What is microtargeting and voter data mining?
  • What is the process and what are the benefits of it?
  • Voter file vs. on a modeled voter file
  • Segmentation and its importance especially in unaffiliated voters it is

Segment #2 – The History of Microtargeting

  • Reflecting on time at the Republican National Convention
  • The 2004 presidential election
  • How the Obama campaign used modeling at a more advanced level

Segment #3 – Microtargeting Overview and Process

  • Difference between modeling survey questions and regular survey questions
  • Why a large survey sample is needed
  • Enhancing files with census data, consumer data, election return data

Segment #4 – Using Modeling to Create Segments

  • Our work in CO, NM and PA for voter segmentation
  • How we use 15 models to put someone in a segment

Segment 5 – Limits of Modeling

  • Why it won’t save a blowout election
  • Why it won’t save a bad candidate
  • Challenges building a Republican primary model