2. Course Introduction
More than 46% of businesses use Business Intelligence tools as a core element of
their strategy
87% low analytics maturity
3. Definition of Analytics
What is Analytics?
Analytics is the discovery, interpretation, and communication of meaning patterns in data1
a. Discovery & interpretation
b. Communication
Non-business analytics
Government studies: GDP
Public Health
Academic Research
1. Reference: https://en.wikipedia.org/wiki/Analytics
4. Art of Analytics
Requires Design knowledge, because data needs to be visualized
Colors
Space
User Interface
Story telling
Requires domain knowledge, to be able to apply findings to real business problems
Requires writing skills to be able to summarize hundreds and thousands of records
in just one headline
5. Data Collection
How is data collected?
Manually through submissions
Benefits: Cheaper – Ease of use – Integrity of the Data – Good Data Privacy and protection – Reliable
– Lack of Technology
Limitation: Time consuming – Storage issues – Loss of Data – Interpretation Error – Time Lag
Automatically through gadgets such as POS systems
Benefits: More data points – Relatively less time lag – easy access – time savings – better accuracy
Limitations: Data privacy/unanimous – over reliance on technology/infrastructure – cost of
infrastructure/technology – potential error in data aggregation – Lack of integrity/Fake Data –
Complex Technology
Semi Automatic
A mix of automated data along with manual adjustments
6. Data Analysis
What is Data Analysis?
It is the process of inspecting, cleansing, transforming a modelling data with the end goal
of discovering useful information and supporting decision making.
2 Types of data analysis
Focuses on What
Quantitative analysis is the process of
collecting and evaluating measurable
data such as revenues to understand the
business performance. It deals with
numerical value
Quantitative Analysis
Focuses on Why
Why do people behave in certain way?
Why do they buy certain item?
It measures feelings, thoughts,
perceptions to understand motivations
and behavior
Qualitative Analysis
8. Descriptive Analytics
Descriptive: What happened?
Is the base of analytics, it’s the foundation, the eldest & what most business do today
Descriptive analytics takes historical data and turns it into digestible chunks
Questions Use Methods Tools
What happened? Which hotel made
more money in the
last 3 years?
-Data aggregations
-Dashboards
-Reports
-Data Analysis
-Excel
-Power BI
-Tableau
-Qlik
-Business Objects
Cognos
Why it happened? Which channel is
contributing to
more reservations?
When it happened?
9. Descriptive analytics: Hotel chain
Your chain have two hotels:
In the City
A resort on the beach
What has happened in the last years with the properties?
What?
0%
20%
40%
60%
80%
City Hotel Resort Hotel
69%
31%
Revenue
Why?
Travel agencies are booking much
more the City hotel
0
1
2
3
4
5
6
7
TA/TO Direct Corporate GDS
Reservation
(bnh)
Resort Hotel
City Hotel
10. Descriptive vs Inferential Statistics
There are 2 types of statistics:
Descriptive
Inferential
Descriptive Statistics helps describe, show or summarize data in a meaningful way.
Example: your company's HR use statistics to calculate the average salary across all
employees
Inferential Statistics allow you to make predictions, to use data samples to make
generalizations about bigger populations.
Example: Government surveys to determine the average salary for the entire nation
11. Types of measures in Statistics
Ways of describing the central position
of a frequency distribution for a group
of data
Examples: Mode, median, and mean
Measures of Central tendency
Ways of summarizing a group of data by
describing how spread out the scores
are
Examples: range, quartile, absolute,
deviation, variance and standard
deviation
Measures of Spread
12. Predictive analytics
Previewing the future
It provides estimates about the likelihood of something to happen in the future
Caveat: Remember that no statistical algorithm can “predict” the future with 100%
certainty
Questions Use Methods Tools
What will happen? How many
reservations will
you get next year?
-Forecasting
-Risk Modeling
-Customer
segmentation
-Sentiment Analysis
-R programming
Data visualization
-IBM SPSS
-Python
When will it happen? Chances of a
customer to
revisit?
13. Predictive Models
What is a predictive model?
Predictive modelling refers to the process of using known results to create,
process and validate a model that can be used to forecast future outcomes.
Examples: Understand customer behavior as well as financial, economic and
market risks.
Applications Infrastructure Tools Best Practices
- Demand
forecasting
- Workforce
planning
- Fleet equipment
maintenance
Data warehouse
Machine Learning
Hadoop
R
Python
Regression
Neural Network
Random Forests
14. Prescriptive Analytics
Prescriptive analytics provides recommendations of different possible actions to
optimize business outcomes
Is the most optimized as well as complex form of analytics because it deals with
many variables
Questions Use Methods Tools
What should I do? What should be
the optimal room
price next year?
-Machine learning
-Algorithms
-Data Modelling
-Alteryx
-Python
-Rapid Minder
-Sisense
How can I make it
happen?
How many rooms
should you
overbook?(consid
ering
cancellations)
15. References
1. Storytelling with data, Cole Nussbaumer Knaflic , 2015
2. Good Leaders Ask Great Questions: by John C. Maxwell
3. Business Intelligence Roadmap, Larissa Terpeluk Moss and S. Atre
Editor's Notes
87% of organizations have low BI and analytics maturity, that means that there’s a plenty of opportunities to improve
Let me provide some backgrounds to the course, we’ll discuss about many matters and principals about the data analytics in this course, and will bring you up to the speed to the conversations that are happening in most of the companies around the world today
Most importantly you’ll learn how to apply data analytics and critical thinking in your future jobs, because this course is focused on business analytics: The art of analysing data to make money
It doesn’t matter if you work in a big corporation or a small start-up, because every single company makes data
Discovery & interpretation: We need to find patterns and do the interpretation of it.
Communication is super important, because it's very important that all the facts, findings that we search through data are properly communicated. Analytics is a vast domain.
When we talk about Analytics, It's important to differentiate two flavors of it: The first one is the non-business analytics.
When we talk about non-business analytics, we speaking, for example, about government studies like
the GDP, public health, academic research or explorative analytics like astrology
Analytics is more than a few lines of coding or just putting together a dashboard or a beautiful visual
A Point of Sale (POS) system allows your business to accept payments from customers and keep track of sales. ...