Professional Documents
Culture Documents
I T I S K N O W L E D G E T H AT L I B E R AT E S
MARKETING
ANALYTICS
Dr. V. Raj Kumar
Assistant Professor
X
Course Objectives
I T I S K N O W L E D G E T H AT L I B E R AT E S
The objective of the course is to show you the benefits of using a systematic and
analytical approach to marketing decision-making. An analytical approach will enable you
to:
1. Use and execute data analytic techniques, and case studies to understand how to
solve marketing analytics problems in a scientific and process-driven manner, and
2. Understand how the “first principles” of marketing strategy help firms organize the
analytics opportunities and challenges that exist in today’s data era.
X
Learning Outcomes
I T I S K N O W L E D G E T H AT L I B E R AT E S
Upon succeeding in this course, students will be able to complete the following tasks:
1. Execute and interpret a factor analysis, cluster analysis, and classification analyses
to better understand how to manage customer heterogeneity by segmenting and
targeting buyers. (L1)
2. Understand how choice model analysis can inform customer preferences and
improve customer lifetime valuation estimates. (L2)
3. Execute and interpret a conjoint analysis study with application to new product
design. (L3)
4. Use a response model to improve marketing resource allocation decisions. (L4)
X Miller, T. W. (2015). Marketing data science: modeling techniques
in predictive analytics with R and Python. FT Press.
I T I S K N O W L E D G E T H AT L I B E R AT E S
TEXT BOOKS Inge, C. (2022). Marketing Metrics: Leverage Analytics and Data
to Optimize Marketing Strategies. Kogan Page Limited.
Grigsby, M. (2022). Marketing analytics: A practical guide to
• Palmatier, R. W., Petersen, J. A., improving consumer insights using data techniques. Kogan Page
Publishers.
Germann, F. (2022). Marketing
Venkatesan, R., Farris, P. W., Wilcox, R. T. (2021). Marketing
Analytics: Based on First Principles. Analytics: Essential Tools for Data-Driven Decisions. United
United Kingdom: Bloomsbury States: University of Virginia Press.
Academic. Verhoef, P. C., Kooge, E., Walk, N., Wieringa, J. E. (2021). Creatin
g Value with Data Analytics in Marketing: Mastering Data
Science. United Kingdom: Taylor & Francis.
Griffel, M., & Guetta, D. (2020). Python for MBAs. Columbia
University Press.
Schwarz, J. S., Chapman, C., & Feit, E. M. (2020). Python for
marketing research and analytics. Springer Nature.
James, G., Witten, D., Hastie, T., Tibshirani, R., & Taylor, J. (2023).
An introduction to statistical learning: With applications in
python. (No Title). https://www.statlearning.com/
Software Tools
MS Excel
Tableau
KNIME
Python
R
•
•
•
•
•
X I T I S K N O W L E D G E T H AT L I B E R AT E S
X
Course Outline
I T I S K N O W L E D G E T H AT L I B E R AT E S
Pedagogy – “Tell-Show-Do”
• Lectures Class Discussion
• Textbook Case Discussion Flipped Classroom
Evaluation Scheme
• Marketing analytics (also called customer analytics, data analytics, or big data) are
changing the very world we live in and how we interface with businesses.
• With marketing analytics, firms can quickly make optimized decisions; they can
connect data in novel ways and thus earn incremental revenue.
• This expansion of marketing analytics has been driven by many technological
advances, such as the internet, GPS, smartphones, cheap data storage and
processing, artificial intelligence (AI), and widespread digitization.
• In many cases, the technology has advanced faster than managers’ ability to apply
the advances effectively.
Analytics
• Data analytics refers to a “multidimensional field that uses mathematics, statistics,
predictive modeling and machine learning techniques to find meaningful patterns and
knowledge in recorded data.”
• Business analytics typically reduces the scope and focuses on data mining,
predictive analytics, applied analytics, and statistics targeted at industry business
processes.
• Big data analytics is similar to data analytics but highlights the amount or breadth of
data processed (i.e., very large amounts) and often the use of specific techniques
(data mining, machine learning, AI) that are most relevant for this type of data
structure.
Marketing analytics can be defined as “a technology-enabled and model-supported
approach to harness customer and market data to enhance marketing decision
making.”
X
Marketing Analytics
Marketing Research Marketing Analytics
Involves only collecting the data that answers Assembles massive amounts of data which tends to
specific questions. be impersonal and multi-faceted.
• Marketing analytics have evolved dramatically due to the increasing amount and availability of
data in the digital economy, as well as novel analysis techniques and uses of data analytics.
• Marketing analytics in turn have evolved in two parallel, interrelated areas: the data and the
analytics.
X
I T I S K N O W L E D G E T H AT L I B E R AT E S
“Hotels can double their profitability by successfully marking up rooms just 10 percent.”
• There are times when there are not enough hotel rooms to meet massive customers'
demand and times with many unused rooms due to low demand
• This is an example ___________ management _______ pricing strategy
• Hotels can vary prices to attempt to match supply and demand in a dynamic fashion
• This is an called _______ pricing strategy
• Using customer data analysis, hotels can drop room prices when demand on the website and
call centers is low but rapidly increase prices when it senses an increase in demand
X
The Data
I T I S K N O W L E D G E T H AT L I B E R AT E S
Science
Process
• SEMMA
• Sample, Explore, Modify, Model,
Assess
• CRISP-DM
• Cross Industry Standard Process
for Data Mining
• KDD
• Knowledge Discovery in
Databases
CRISP-DM
X I T I S K N O W L E D G E T H AT L I B E R AT E S
The Team Data Science Process (TDSP)
X I T I S K N O W L E D G E T H AT L I B E R AT E S
X I T I S K N O W L E D G E T H AT L I B E R AT E S
X
• https://www.youtube.com/watch?v=OdW7vhYYSdM
• Watch this video and describe what they are doing with reference to CRISP-DM
process (see previous slide)
X
I T I S K N O W L E D G E T H AT L I B E R AT E S
Data Model
Model Training Model Testing Deployment
Preparation Optimization
Data Manipulation Model Training Parameter Tuning Performance Files & DBs
Data Blending Bag of Models Parameter Optimization Measures Dashboards
Missing Values Handling Model Selection Regularization Accuracy REST API
Feature Generation Ensemble Models Model Size ROC Curve SQL Code Export
Dimensionality Reduction Own Ensemble Model No. Iterations Cross-Validation Reporting
Feature Selection External Models … … …
Outlier Removal Import Existing Models
Normalization Model Factory
Partitioning …
…
Any Questions?
21
X I T I S K N O W L E D G E T H AT L I B E R AT E S