Lecture 1 - Course introduction
Course Digital and social media strategies
Status Done
Tags
Semester Semester 5
Type Lecture
Files week_01_slides.pdf
Learning goals
Develop an understanding what’s needed in a Digital Marketer’s toolkit
Digital and Social Media Strategy: A
possibly biased perspective
What is digital marketing
It is the marketing of products or services using digital technologies (mainly internet, but
also mobile phones, display ads and any other digital media)
What is social media marketing
It is the process of creating (tailored) content to drive engagement, promote a business
and generate revenue.
What is marketing strategy
A firm’s overall game plan for reaching prospective consumers and turning them into
customers for their products or services.
Methods to analyze data
3 main ways to analyze data
Lecture 1 - Course introduction 1
, 1. Descriptive analysis (summarize characteristics of a data set) - we’re doing
some of this
What does the data look like?
Means, standard deviations distribution of data
Results are (stylized) facts
Examples
How are users who discuss a movie connected on social media?
Is revenue attributed to clicks from search engines higher than last week?
Has revenue growth from digital stores stayed high after COVID-19?
2. Causal analysis (does A lead to B?) - we’ll spend a lot of time in this space
Might also care about the mechanism of how it happens
Examples
Do search engine ads increase product purchases?
Does product promotion by influencers increase demand?
Do push notifications work? In what way?
3. Predictive analysis (how can I best predict an outcome?) - not focus of this
class
When A occurs, so does B
Examples
Is a review posted by a real person or by a bot?
How many retweets does Nike expects its next tweet to get?
Who is most likely to click on a digital ad?
Lecture 1 - Course introduction 2
, How to deliver data driven answers?
Goal is to deliver data driven insights to strategic questions. You need to develop an
analytics toolkit (mainly statistical, building on existing knowledge)
Good news = high quality data driven solutions are incredibly useful. Why?
D&SM impacts a wide variety of industries
Digital Markets deliver a deluge of data
It provides real answers to real problems in marketing and business strategy
Being able to do (good) strategy relevant analytics ensures many (enjoyable)
job prospects
Bad news = it’s hard
Need to learn how to use the right tools & be able to work on causal (and
predictive) questions, while understanding the context of the problem at hand
Learning tools can be boring
Lecture 1 - Course introduction 3
, Lecture 2 - Experiments
Course Digital and social media strategies
Status Done
Tags
Semester Semester 5
Type Lecture
Files
Learning goals
Translate strategic questions into testable hypotheses
Explain the difference between good data-driven analysis and traditional data-driven
analysis
Explain why correlation is not (generally) causation
Identify limitations of working with observational data
Discuss how experimental data overcomes limitations of observational data
Compute descriptive statistics and visualize experimental data
Use statistical hypothesis testing to analyze experimental data and to answer a
strategic question
Causal Data Driven Marketing
What is the impact of an
marketing intervention (X) on an
outcome (Y)
Hard to evaluate
Lecture 2 - Experiments 1
Course Digital and social media strategies
Status Done
Tags
Semester Semester 5
Type Lecture
Files week_01_slides.pdf
Learning goals
Develop an understanding what’s needed in a Digital Marketer’s toolkit
Digital and Social Media Strategy: A
possibly biased perspective
What is digital marketing
It is the marketing of products or services using digital technologies (mainly internet, but
also mobile phones, display ads and any other digital media)
What is social media marketing
It is the process of creating (tailored) content to drive engagement, promote a business
and generate revenue.
What is marketing strategy
A firm’s overall game plan for reaching prospective consumers and turning them into
customers for their products or services.
Methods to analyze data
3 main ways to analyze data
Lecture 1 - Course introduction 1
, 1. Descriptive analysis (summarize characteristics of a data set) - we’re doing
some of this
What does the data look like?
Means, standard deviations distribution of data
Results are (stylized) facts
Examples
How are users who discuss a movie connected on social media?
Is revenue attributed to clicks from search engines higher than last week?
Has revenue growth from digital stores stayed high after COVID-19?
2. Causal analysis (does A lead to B?) - we’ll spend a lot of time in this space
Might also care about the mechanism of how it happens
Examples
Do search engine ads increase product purchases?
Does product promotion by influencers increase demand?
Do push notifications work? In what way?
3. Predictive analysis (how can I best predict an outcome?) - not focus of this
class
When A occurs, so does B
Examples
Is a review posted by a real person or by a bot?
How many retweets does Nike expects its next tweet to get?
Who is most likely to click on a digital ad?
Lecture 1 - Course introduction 2
, How to deliver data driven answers?
Goal is to deliver data driven insights to strategic questions. You need to develop an
analytics toolkit (mainly statistical, building on existing knowledge)
Good news = high quality data driven solutions are incredibly useful. Why?
D&SM impacts a wide variety of industries
Digital Markets deliver a deluge of data
It provides real answers to real problems in marketing and business strategy
Being able to do (good) strategy relevant analytics ensures many (enjoyable)
job prospects
Bad news = it’s hard
Need to learn how to use the right tools & be able to work on causal (and
predictive) questions, while understanding the context of the problem at hand
Learning tools can be boring
Lecture 1 - Course introduction 3
, Lecture 2 - Experiments
Course Digital and social media strategies
Status Done
Tags
Semester Semester 5
Type Lecture
Files
Learning goals
Translate strategic questions into testable hypotheses
Explain the difference between good data-driven analysis and traditional data-driven
analysis
Explain why correlation is not (generally) causation
Identify limitations of working with observational data
Discuss how experimental data overcomes limitations of observational data
Compute descriptive statistics and visualize experimental data
Use statistical hypothesis testing to analyze experimental data and to answer a
strategic question
Causal Data Driven Marketing
What is the impact of an
marketing intervention (X) on an
outcome (Y)
Hard to evaluate
Lecture 2 - Experiments 1
