@hyunjeehale
The study of the distribution and determinants of health-related states and events in specified populations
John Snow
One of the Fathers of Modern Epidemiology
1854 Broad Street
Cholera Outbreak London
Where Do Epidemiologists Work?
Digital Campaigns vs.
Disease
The Epidemiologic Triangle
Environment
Host
Agent
The Digital Triangle
Digital Presence
Target Audience
CTAs
Natural Hx of Disease
Primary Prevention Secondary Prevention
Interrelations of Agent, Host, and Environmental Factors
Rxn of the Host to the Stimulus
Levels of Preventive Measures
Tertiary Prevention
RehabilitationDisability Limitation
Early Dx & Prompt Tx
Specific Protection
Health Promotion
Prepathogenesis Period Period of Pathogenesis
Production of Stimulus Early Pathogenesis
Discernible Early
Lesions
Advanced Disease
Convalescence
Natural Hx of Digital
Primary Promotion Secondary Promotion
Interrelations of Target Audience, Digital Content & Environment
Rxn of the Target Audience to the Digital Campaign
Levels of Promotion
Tertiary Promotion
Maintenance
Paid AdvertisingAudience Collection
Content Marketing, SEO
& Native Advertising
Receptive, Undecided Path to Conversion & Retention
Building Brand Awareness Interest & Consideration
Actively Seeking
ConversionRetention &
Referrals
Web of Causation
Web of Digital
Epidemiologic Study Design
Observational
Cross-Sectional
AnalyticDescriptive
Case-Control
Cohort
Digital Study Design
Observational
Analyze all visitors to a website and determine
exposure to the campaign and
conversion rates
AnalyticTarget
Audience
Separate visitors into
convertors and non-convertors and determine
exposure
Follow visitors who were exposed to a
campaign vs. visitors who were
not and determine rate of conversion
Epidemiologic Study Design
Experimental
Community Trials
Clinical Trials
Quasi-Experimental
Digital Study Design
Experimental
Change specific features of the digital
campaign
Random A/B testing of landing pages
Focus Groups
Statistical Decision-Making
Question of Interest
Sample/Population
Study Design
Data Collection
Description of the Data
Formulate Hypothesis
Inferential Statistics
Interpretation of Results
Additional Studies/Implement Program Initiatives
Digital Decision-Making
Business Objective
Target Audience
Marketing Plan
Data Collection
Description of the Data
Visitors are more likely to convert when X
Test if your theory is correct using statistical
models
Reporting to internal teams and clients
Develop modified digital strategy based on results
Incidence and Prevalence
incidence
prevalence
New and Existing Customers
new
existing
bathtub(digital landscape)
preventsdrop-off
existing
retainlose
1,000 Mortality
Rate
100 Conversion
Rate
1,000 Mortality Rate
100 Conversion
Rate
Years of Potential Life Lost (YPLL)
Life Expectancy
Age at Death
Years of Customer Lifetime Lost
Avg. Customer Lifespan
Number of Years a Customer at time
of Loss
Measure for Assessment of Risk
Absent
TOTAL
b
Exposure
Present
Absent
Disease
Present
a
dc
TOTAL
a+c b+d
a+b
c+d
a+b+c+d
Measure for Assessment of Conversion
Absent
TOTAL
b
Exposure
Present
Absent
Conversion
Present
a
dc
TOTAL
a+c b+d
a+b
c+d
a+b+c+d
What’s Next?
Who Do You Need?
Coming in 2017 Data Detectives & Digital Space
Finding the Signals within the Noise