Business Analytics
CS Elective course
Crouse Plan
Prepared by Dr. Rauf Ahmed Shams Malick (Rauf.malick@nu.edu.pk)
Motivation
Business analytics is an integral part of modern management. This course will provide students with best practices and
tools to identify value capture as well as understand and apply state-of-the art methods to drive value. The course will be
executed in an activity-based problem solving fashion.
The ability to use data effectively to drive rapid, precise, and profitable decisions has been a critical strategic advantage
for companies as diverse as Zillow, Careem, Systems, Daraz and Affinity. In addition, many current and recent startups are
progressing locally to thrive in the age of analytics. With the increasing availability of broad and deep sources of
information – so-called “Big Data” – business analytics are becoming an even more critical capability for enterprises of all
types and all sizes.
In this course we will experiencing the data-intelligence driven strategies and hands on problem solving on data-oriented
business cases. Course will extensively demand hands on involvement of machine learning tools, exploratory data analysis
and analytical skills throughout the entire duration.
After the course, you will be able to perform the following:
i.
ii.
iii.
iv.
v.
vi.
vii.
viii.
ix.
x.
xi.
xii.
xiii.
xiv.
xv.
Design, application development and management of Data Engineering pipelines for web and mobile
applications.
Design and establishment of web analytics through multiple tools including Google Analytics, Piwiki, Pixel.
Design and integration of mobile analytics including heat maps and location based analytics.
Analyses framework for customer insights along with design and implementation of customer pipeline for
web and mobile applications.
Ability to develop customer analytics framework including customer life time value (through data), customer
churn prediction, and customer segmentation.
Ability to design predictive model for customer journey and development of funnels.
Complete data to predictive models’ development for customer credit ranking problem (with the help of data
driven practical case and local startup seminar)
Establishment of customer journey for online retail stores including Daraz and Kahzane.
Workshop from customer journey and experience department from Careem. This workshop will help you in
designing your own pipeline accordingly
Development of Key Performance Indicators KPI for product life time including operations and transactions
Design and development of KPIs for customer journey, examples for Careem, Airbnb, Daraz.
Development and integration of recommendation engine for digital platforms
Seminar from Data Science team from industry including Affinity and HBL.
Ability to understand and design predictive modeling and perspective modeling
Understanding and campaign management for digital advertisement campaign over mobile and web.
Understanding and integration of social media analytics including Facebook, Youtube, and Instagram.
xvi.
Development of Business Dashboards through PowerBI.
Expected project mentorship from Business Analytics leaders from companies including Etilize, Avanza, CoreData, Affiniti,
HBL and Careem.
Course Outline
Following are the core components of the course
Week One
i.
ii.
Data Pipeline vs Analytics Pipeline
Data curation methods and exercises (R or Python)
Week Two
i.
ii.
iii.
Analytics 3.0 and Business Development Strategies.
Online presentations of analytics’ problem faced by the industry including Careem and Etilize.
Introduction to PowerBI
Class activity and demonstration of
Case 1: Bank’s Credit Ranking System (Data Oriented) A data-oriented case, you will be provided a detailed scenario with
data, several machine learning models will be developed and comparison will be performed.
Week Three & Four User Experience, Usability, and Web Analytics
Web analytics help you collect and analyze activities on your website. It helps you figure out things like where your
visitors came from, what they do on your site, how long they stay, the content they like, and so much more.
i.
ii.
iii.
iv.
Understanding product usability, experience and customer journey
Analytics for experience and customer journey (How to interpret the Google Analytics features with
customer journey and to elaborate the user experience)
Key performance indicators for user experience
Understanding, and integration of Pixel, Piwiki, and Google analytics with web-based applications.
Week Five & Seven Customer Journey & Pipeline (Careem & Daraz)
Understanding the customer journey and development of customer data pipeline. Understanding the design of
customer funnel. Digital advertisement and customer funnel.
Development of customer analytics based KPIs. Significant features have to be identified from customer perspective
from web analytics datasets. Real time data will be provided to model the right KPI and predictive models will be
developed subsequently.
i.
ii.
Customer Journey @ Careem and user experience (Careem team seminar)
Mobile Analytics tools
Week 6th Mid 1
Week Eight Online Retail Analytics (Marketplace, B2B, B2C Systems)
Understanding B2B, B2C, and Marketplace. Customer Analytics for each type. Product pricing models will be developed
for online retail stores along with predictive models for profit maximization. Design of customer Journey for Daraz
(Marketplace), and Zillow (like Zameen in Pakistan)
Week Ninth Social Media Analytics (Data Oriented)
Social media KPIs will be developed for Twitter, Facebook and Youtube for multiple business scenarios. Real time data
will be provided for predictive models and KPIs development. Advertisement and marketing strategies.
Week Ten: Recommender Systems
Recommender systems will be discussed along with product recommendation exercise for e-commerce platform. Item
based recommendations and collaborative filtering. Draz case will be discussed.
Week 11th Mid 2
Week Twelve – Knowledge Graphs
Understanding the graph databases and knowledge graphs. Design and implementation of knowledge graphs.
Implementation through Neo4j graph based engine. Knowledge Graphs based recommendations.
Week Thirteen & Fourteen Project Presentation
Students’ presentations, discussions on cases and formulation of problems, projects, services and products will be the
core essence of the course.
Software Tools:
PowerBI, Google Analytics, Piwiki, Facebook Pixel, Neo4j
Course Methodology:
The course will be covered on case-based methodology. During the semester there will be a coverage of at least five cases.
Data intensive cases will be utilized from Harvard and MIT repositories along with a few local cases. Students will be
provided the case document along with the data and they have to perform following against every case before the session.
Case Study Steps (two weeks for each case):
a.
b.
c.
d.
e.
Understanding and elaborating the problem
Understanding the data and extensive utilization of statistical methods for data valuation
Development of Analytics matric based on digital strategies and published articles
Development of predictive models and utilization of NLP and social network analytics methods.
Development of a comprehensive case report.
Each case will be covered in two weeks period by completing the above steps. Each class will be dedicated to respective
stage of the case study. Heavy class participation is desired and will be rewarded accordingly.
Predictive Modeling techniques, computational linguistics methods, social network analysis, and other tools will be
employed in building models on real world data oriented problems. Data from Google Analytics, social media and
customer choices will be used to understand the customer and to build the predictive models for several problems.
Case studies will be studied and experiments will be performed in the following domains (not restricted to):
-
Digital Retail (e-Commerce platforms)
Sales prediction data
Credit value determination problem
Social Media Datasets
Course Evaluations:
Along with the two sessional exams (one in case of MS) each case step (as mentioned in section 2) will be rewarded in
terms of assignments and quizzes. Class participation is highly desired thereby highly rewarded. A course project will be
submitted by the students by the end of the 12th week of the course. During the project each student have to develop a
business problem along with the supported dataset by his own. All learned methodologies have to be employed for
problem solving. By the end of the course (during the last 4 weeks) students will be presenting their cases in details in
front of students.
Marks Distribution:
Sessional
Project:
Case Assignments:
Final:
25 Marks
15 Marks
20 Marks
40 Marks