Marketing analytics*
A.A. 2023/2024
Obiettivi formativi
Familiarize students with data-driven marketing strategies and to help them understand the process of converting data to marketing decisions. Provide a working knowledge of data handling and modeling techniques using widely-used software and tools. Present applications of the techniques to new product design, consumer segmentation, targeting customers, pricing, search engine advertising, and sales force allocation. Acquire a practical knowledge of the use of big data and of machine learning and statistical learning techniques in marketing analytics
Risultati apprendimento attesi
At the end of the course the student will feel comfortable making data-driven marketing decisions independently as well as in a group setting, will gain competency to utilize the commonly-used software tools for his / her marketing-related data analysis needs, will be able to carry out these techniques independently after practicing on several full-length cases.
Periodo: Primo trimestre
Modalità di valutazione: Esame
Giudizio di valutazione: voto verbalizzato in trentesimi
Corso singolo
Questo insegnamento può essere seguito come corso singolo.
Programma e organizzazione didattica
Edizione unica
Responsabile
Periodo
Primo trimestre
Programma
Marketing essentials
· Segments and markets
· Competitive benchmarking
· Customer problem and Value proposition
· Customer choice and decision making bias
· Customer journey analysis
· Marketing KPIs: CAC Customer Acquisition Cost. LTV Lifetime Value, ROMI Return on Marketing Investment
· Funnel metrics: conversion rate, retention rate, churn rate
· Sources of marketing data and Marketing data models: structured, semi-structured and unstructured data
Segmentation and targeting
· STP framework (Segmentation, Targeting, Positioning)
· Traditional segmentation
· Segmentation via clustering: K-means
· Hierarchical clustering
· Behavioral segmentation
· RFM Recency Frequency Monetary analysis
· Targeting and resource allocation
· GE Matrix
Positioning
· Attribute-based product analysis
· PCA Principal Component Analysis
· Perceptual maps, Preference maps and Joint-space maps
· Reverse mapping
Customer Lifetime Value
· Customer value assessment
· Customer valuation
· Customer profitability
· Customer lifetime value modeling
· Predicting customer revenue
· Predicting customer churn
New Product Design
· New product development
· Conjoint Analysis
· Collecting data for conjoint analysis
· Product design options, Attribute levels and bundles
Marketing mix and resource allocation
· Marketing mix
· Pricing
· Response models
· Resource allocation
Digital Marketing
· Web analytics
· Online advertising
· Search advertising analytics
· Panel data
· Segments and markets
· Competitive benchmarking
· Customer problem and Value proposition
· Customer choice and decision making bias
· Customer journey analysis
· Marketing KPIs: CAC Customer Acquisition Cost. LTV Lifetime Value, ROMI Return on Marketing Investment
· Funnel metrics: conversion rate, retention rate, churn rate
· Sources of marketing data and Marketing data models: structured, semi-structured and unstructured data
Segmentation and targeting
· STP framework (Segmentation, Targeting, Positioning)
· Traditional segmentation
· Segmentation via clustering: K-means
· Hierarchical clustering
· Behavioral segmentation
· RFM Recency Frequency Monetary analysis
· Targeting and resource allocation
· GE Matrix
Positioning
· Attribute-based product analysis
· PCA Principal Component Analysis
· Perceptual maps, Preference maps and Joint-space maps
· Reverse mapping
Customer Lifetime Value
· Customer value assessment
· Customer valuation
· Customer profitability
· Customer lifetime value modeling
· Predicting customer revenue
· Predicting customer churn
New Product Design
· New product development
· Conjoint Analysis
· Collecting data for conjoint analysis
· Product design options, Attribute levels and bundles
Marketing mix and resource allocation
· Marketing mix
· Pricing
· Response models
· Resource allocation
Digital Marketing
· Web analytics
· Online advertising
· Search advertising analytics
· Panel data
Prerequisiti
The course assumes prior knowledge of statistics, machine learning concepts and methods, as well as Python coding skills with particular regard to data manipulation, visualization and modeling. Some of these skills will be briefly reviewed in class but a good starting working knowledge is strongly advised. These skills are especially needed in order to be able to attend effectively the practice sessions, the study cases and the Capstone.
Metodi didattici
Course topics will be discussed first in theoretical terms, while keeping the discussion as open as possible and based on readings, articles, videos and other teaching devices in order to maximize participation. Most theoretical topics will then be featured in practical lectures where case studies and real-world datasets will be used to provide students with an active learning environment and experience.
Practical activities will be undertaken in groups, formed during the first lecture or shortly thereafter. At the end of each practical session, one or two groups will be randomly selected to present their solution, analyses and deliverables related to the case to the rest of the class. Groups not asked to present will have to still be actively engaged in the discussion.
Practical activities will be undertaken in groups, formed during the first lecture or shortly thereafter. At the end of each practical session, one or two groups will be randomly selected to present their solution, analyses and deliverables related to the case to the rest of the class. Groups not asked to present will have to still be actively engaged in the discussion.
Materiale di riferimento
Course textbook: Gary L. Lilien, Arvind Rangaswamy, Arnaud De Bruyn - Principles of Marketing Engineering and Analytics, 3rd edition, DecisionPro Inc, ISBN-13: 978-0985764821
Additional materials (slides, exercises, exam simulations) in the ARIEL website. Case study datasets will be made available either via the aforementioned platform or via specific methods (API access, SQL access, Github, shared cloud resources, URLs, etc.) with instructions provided where needed.
Additional materials (slides, exercises, exam simulations) in the ARIEL website. Case study datasets will be made available either via the aforementioned platform or via specific methods (API access, SQL access, Github, shared cloud resources, URLs, etc.) with instructions provided where needed.
Modalità di verifica dell’apprendimento e criteri di valutazione
Participation
The course will be a combination of lectures, discussions, practical activities with case study datasets as well as group work tasks and presentations. Students are expected to take an active role in class discussions as well as to come prepared to class, having completed readings and other assignments indicated for each lecture. Active participation will be part of the final evaluation.
Capstone/Project Work
Throughout the course, groups will be engaged with a real world marketing analytics endeavor, related to external companies, partnering with the course and the instructor. In addition to the time in class, groups will have to allocate some time for additional effort outside the course schedule. A final deliverable, in the form of a written report will have to be submitted electronically as part of the final exam. The evaluation for this deliverable will be formed by a group evaluation, combined with a peer evaluation whereby students evaluate each other (respectively, 60% and 40% of the final Project Work evaluation).
Exam instructions
Both for attending and non-attending students, the Project Work constitutes the sole required step, considering the complexity of the task itself. As mentioned above, the outcome of the Project Work takes the form of a written report, including text, code and any necessary exhibit/attachment. This package is to be sent via email to the instructor within a deadline agreed upon when setting up the Project Work (see below). This email delivery completes the exam requirements, i.e. no public/private presentation will be required (though it could still be requested by the students, in order to be able to better communicate the value of the work conducted, and, in that case, a remote Teams session could be organized).
The procedure to setup the Project Work consists in: (1) choosing the topic, among the ones proposed by the instructor; (2) setting up a deadline that takes into account the complexity of the task and the number of people in the PW group; (3) setting any additional details specific to the task at hand, if necessary. This procedure to setup the Project Work necessarily differentiates between attending and non-attending students.
For attending students, Project Work groups will be formed in class, following teacher's instructions and specific deadlines and methods communicated during the course.
Non-attending students will have to contact the instructor in advance to receive instructions as well as one or more proposals for the Project Work's topic. Given the workload associated to the Project Work, it is advisable to form groups (even small ones) also in the case of non-attending students. The deadline to contact the instructor in order to setup the Project Work as a non-attending student or student group is 4 weeks before the desired official exam date, among the ones published on the University portal. Any request received later than that may have to be postponed to the subsequent official exam date. The email to indicate your intention to setup a Project Work as a non-attending student will have to report the following subject "Marketing Analytics non-attendant PW request".
For both attending and non-attending students, please note that the deadline to deliver the final PW report does not necessarily match the date of the exam for which the students registered ("Appello"). Therefore, if you need more time due to job schedule etc. this can be easily accommodated.
Finally, for the first exam dates in December 2023/January 2024, non-attending students have had the opportunity to arrange for slightly different arrangements for the Project Work. Therefore, if you have already setup a PW group, the instruction above do not apply to you.
The course will be a combination of lectures, discussions, practical activities with case study datasets as well as group work tasks and presentations. Students are expected to take an active role in class discussions as well as to come prepared to class, having completed readings and other assignments indicated for each lecture. Active participation will be part of the final evaluation.
Capstone/Project Work
Throughout the course, groups will be engaged with a real world marketing analytics endeavor, related to external companies, partnering with the course and the instructor. In addition to the time in class, groups will have to allocate some time for additional effort outside the course schedule. A final deliverable, in the form of a written report will have to be submitted electronically as part of the final exam. The evaluation for this deliverable will be formed by a group evaluation, combined with a peer evaluation whereby students evaluate each other (respectively, 60% and 40% of the final Project Work evaluation).
Exam instructions
Both for attending and non-attending students, the Project Work constitutes the sole required step, considering the complexity of the task itself. As mentioned above, the outcome of the Project Work takes the form of a written report, including text, code and any necessary exhibit/attachment. This package is to be sent via email to the instructor within a deadline agreed upon when setting up the Project Work (see below). This email delivery completes the exam requirements, i.e. no public/private presentation will be required (though it could still be requested by the students, in order to be able to better communicate the value of the work conducted, and, in that case, a remote Teams session could be organized).
The procedure to setup the Project Work consists in: (1) choosing the topic, among the ones proposed by the instructor; (2) setting up a deadline that takes into account the complexity of the task and the number of people in the PW group; (3) setting any additional details specific to the task at hand, if necessary. This procedure to setup the Project Work necessarily differentiates between attending and non-attending students.
For attending students, Project Work groups will be formed in class, following teacher's instructions and specific deadlines and methods communicated during the course.
Non-attending students will have to contact the instructor in advance to receive instructions as well as one or more proposals for the Project Work's topic. Given the workload associated to the Project Work, it is advisable to form groups (even small ones) also in the case of non-attending students. The deadline to contact the instructor in order to setup the Project Work as a non-attending student or student group is 4 weeks before the desired official exam date, among the ones published on the University portal. Any request received later than that may have to be postponed to the subsequent official exam date. The email to indicate your intention to setup a Project Work as a non-attending student will have to report the following subject "Marketing Analytics non-attendant PW request".
For both attending and non-attending students, please note that the deadline to deliver the final PW report does not necessarily match the date of the exam for which the students registered ("Appello"). Therefore, if you need more time due to job schedule etc. this can be easily accommodated.
Finally, for the first exam dates in December 2023/January 2024, non-attending students have had the opportunity to arrange for slightly different arrangements for the Project Work. Therefore, if you have already setup a PW group, the instruction above do not apply to you.
Docente/i