MICRO-TARGETING AND INTERNET TESTED MAIL 1
Dec 05, 2014
MICRO-TARGETING AND INTERNET TESTED MAIL
1
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
TARGETING BEGINS WITH DATA
3
• Voter File- the universal base file• Individual Demographics – Age,
Race, Gender, Marital Status, Income
• Geography- Census Data• Precinct results (Electoral Zones)• Vote History- Vote in primaries,
generals• Voter ID/Party ID • Enhancements – Ethnic surnames,
Phone match• Lifestyle (Psychographics)-
Hunter/fisher, veteran, churchgoer, union member, consumer buying power
GEOGRAPHIC TARGETING HELPSbut has its limits
4
• Census Data• Precinct results
(Electoral Zones)• Neighborhood
culture
Birds of a Feather Flock Together
Micro-Targeting allows you to target the individual or household,
not just the neighborhood
5
The Difference Between Polling and Micro-Targeting
6
Polling• Small sample
– 800-1200 interviews
• Long interview –
70-100 questions
over 20-25 minutes
• Limited two- dimensional cross-tabulations:
Women over age 50
College Educated Men Unmarried women
Micro-Targeting• Huge sample – 8,000-12,000 interviews
• Short interview - 8 questions maximum
• Multi-dimension cross-tabulations. 300
variables x 300 variables
Micro-Targeting combines issue questions with voter file data
7
• Voter File- the universal base file• Individual Demographics – Age, Race,
Gender, Marital Status, Income• Geography- Census Data• Precinct results (Electoral Zones)• Vote History- Vote in primaries,
generals• Voter ID/Party ID • Enhancements – Ethnic surnames,
Phone match• Lifestyle (Psychographics)-
Hunter/fisher, veteran, churchgoer, union member
1. What party do you most indentify with?
2. Agree/disagree with following statement: Health Care system is broken and needs major reform
3. If the election were held today would you vote for McCain or Obama?
4. How strongly do you support that candidate?
5. How often do you attend church or religious services?
6. Are you or other members of your family active duty military, veteran or member of the National Guard or Reserves
MODELING INDENTIFIES YOUR INDIVIDUAL TARGETS
• Scores every voter on the file from 0-100 – divides into 10 equal buckets ranging from most Pro-McCain to most Pro-Obama
• Models clearly identify persuasion and GOTV targets
8
Response Rate by Score Decile
0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
1.2%
1.4%
1.6%
1-10% 11-20% 21-30% 31-40% 41-50% 51-60% 61-70% 71-80% 81-90% 91-100%
Response Score
Res
pons
e Rat
e
10987654321
Highest Scores, Most Pro-Obama
Lowest Scores Pro-McCain
MULTI-DIMENSIONAL MODELING GOES BEYOND GEOGRAPHY AND CLASS
• The models generate scores that can be used to – Identify Persuadable voters– Identify Issue voters– Identify GOTV voters
• This approach has successfully helped Republicans find their niche voters and now Democrats are catching up, and are using modeling to find donors as well as voters.
9
TWO HOUSES, SAME DEMOGRAPHICS
• Votes Democratic• Union Member
• Votes Republican • Weekly Churchgoer• Hunting License
10
Same Age, Income, Marital Status, Race, # of Children What makes them different?
LEFT RIGHT
WHAT IF YOU ONLY HAVE LIMITED LIFESTYLE INFO?
• Build a statistical model that compares your list of supporters with the entire voter universe.
• Create an algorithm that scores every voters 1-100 with their likelihood to look and act like your supporters.
11
TWO HOUSES, SAME DEMOGRAPHICS
• Leans Democratic
Model Score = 56
• Votes Republican • Hunting License
Model Score = 32
12
Same Age, Income, Marital Status, Race, # of Children What makes them different?
LEFT RIGHT
African-American
GOTV
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
In
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
USING THE INTERNET TO TEST MAIL
18
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y
B E H I N D E V E R Y V I C T O R Y T H E R E I S A S T R A T E G Y