In this course you will learn how to create models for decision making. We will start with cluster analysis, a technique for data reduction that is very useful in market segmentation. You will then learn the basics of Monte Carlo simulation that will help you model the uncertainty that is prevalent in many business decisions. A key element of decision making is to identify the best course of action. Since businesses problems often have too many alternative solutions, you will learn how optimization can help you identify the best option. What is really exciting about this course is that you won’t need to know a computer language or advanced statistics to learn about these predictive and prescriptive analytic models. The Analytic Solver Platform and basic knowledge of Excel is all you’ll need. Learners participating in assignments will be able to get free access to the Analytic Solver Platform.
0. What is Cluster Analysis
Loading...
From the course by University of Colorado Boulder
Business Analytics for Decision Making
183 ratings
Explore Related Videos
At Coursera, you will find the best lectures in the world. Here are some of our personalized recommendations for you
University of Colorado Boulder
183 ratings
Course 3 of 5 in the Specialization Advanced Business Analytics
From the lesson
Data Exploration and Reduction — Cluster Analysis
Meet the Instructors
Manuel LagunaProfessor
Leeds School of Business
Welcome to the first module of using analytics to improve decision making
in business.
This is the third course in the introduction to business analytics
specialization.
In this course, we transition from the descriptive and
predictive models that we introduce in the second course,
to the prescriptive models that we will discuss here.
We start this transition by answering the question, what is cluster analysis?
As discussed in the two previous courses, there are three types of analytics models,
descriptive, predictive, and prescriptive.
Cluster analysis is one of those, so called, data mining tools.
These tools are typically considered predictive, but since they help managers
make better decisions, they can also be considered prescriptive.
The boundaries between descriptive, predictive and
prescriptive analytics are not precise.
However, as we move from descriptive to predictive and then prescriptive models,
it is fair to say that the analysis tends to be a bit more complex.
But it's very useful if it is done well, cluster analysis refers to a technique for
identifying groups in which the elements of each group are similar.
The approach is often used in marketing in order to understand differences
among customers.
The main idea of this application of cluster analysis
is to create homogeneous market segments to target product offerings.
In this graphical representation we show how in practice is not possible to find
perfectly homogeneous groups.
However, the groups resulting from cluster analysis are similar in some way.
For example, the orange cluster includes round objects.
Not all of them are of the same color or size, and
some are crossed out, but they are all round.
The objects in the green and blue clusters are also similar among themselves,
but they are not exactly the same.
What is important is to be able to describe each cluster.
In this simple example,
the orange cluster may be described as a group of multi-colored round objects.
The green cluster may be characterized as a group of square objects.
And the blue cluster consists of triangular and diamond-shaped objects.
Now, let's take a look at a real world example.
Consumers around the world continue to embrace natural and organic products.
Consumers have become increasingly concerned
about what is in their products they buy.
They have created an industry that, in 2012, the market intelligence firm,
Information Resources Inc., estimated to be $81 billion in the United States.
Information resources survey 5,000 customers and
match the participants answers with the actual purchasing behavior.
The analysis of the data focus on segmenting the US market
by considering relationships between shoppers and
natural, organic, and eco-friendly products.
The market segmentation was assigned to help manufacturers and
retailers to achieve new levels of growth
through a better understanding of these shoppers needs.
The cluster analysis identified seven consumer segments based on lifestyle,
history of purchasing organic and
natural products, attitudes towards these products, and demographics.
These segments were labeled as True Believers, Enlightened Environmentalist,
Strapped Seekers, Healthy Realists, Indifferent Traditionalists,
Struggling Switchers and Resistant Non-believers.
Detecting the segments that are good targets for
natural and organic markets was one of the main outcomes of the analysis.
True Believers and
Enlightened Environmentalists were identified as a key segments.
They represent 18% of the population but together,
they drive nearly half of the total sales.
True Believers are passionate about staying fit and healthy, they are focused
on trying new things and serving as a strong role models for their children.
They are strong believers in the benefit of natural and organic products,
their median income is highest among all segments.
Their average age is 40 and they have at least a college education.
Enlightened Environmentalists are passionate about the environment and
about making good choices to preserve it.
These shoppers are committed to make healthier choices and
go out of their way to shop at stores that carry natural and organic products.
Enlightened Environmentalist are older than True Believers
averaging 63 years old, but have a slightly lower median income.
It is interesting to note that the analysis show that both of these segments
shared the same top priorities regarding food, home care, and personal care.
The importance of this cluster analysis,
is that it was able to identify groups of manufacturers and
retailers who can motivate to learn more about organic and natural products.
For example, the analysis show that the Strapped Seekers and
Healthy Realist are a strong target, driving 24% of sales,
or representing 25% of the consumers.
Strapped Seekers believe and seek natural products, when their budget allows it.
They realize the benefit of natural and organic products and
have the potential of becoming True Believers as their income increases.
Healthy Realists are passionate about being fit but
find it difficult deciding whether to buy healthy or traditional products.
Closer analysis show that not all segments are worth pursuing.
Indifferent Traditionalists lead a simple life with few passions.
They are not likely to buy natural/organic products because
they don't see the reason to change.
Struggling Switchers know that they should be eating healthier and
exercising more but focus on staying within their budgets.
Resistant Non-believers got very little desire to support all their options for
things to buy, so they stay loyal to the products they know.
Market segmentation is not the only application of cluster
analysis in business.
Clustering is also used in market strategies such as product and
price segmentation.
Because applications in areas such as human resources and operations management.
In product segmentation, closer analysis identifies groups of consumers
who want different benefits or levels of functionality of a product category.
The results are used to offer different versions of a product,
in order to match consumers needs more precisely.
Coffee producers, for example, offer regular and
decaffeinated versions of the same brand.
In brand segmentation the same product is offered at different prices
to different groups.
Higher prices are offered to customers in groups that are willing to pay for extras
or pay for better quality packaging, or for additional customer services.
In human resources, clustering could identify characteristics of successful
employees in order to improve recruiting and hiring practices.
Grouping customers by geographical areas is a typical application
of cluster analysis in operations management.
We have shown how cluster analysis can be very useful as a market segmentation tool.
In this module, we will explore additional clustering applications, but first,
we will learn how to perform the analysis and how to interpret results.
Explore our Catalog
Join for free and get personalized recommendations, updates and offers.
Coursera provides universal access to the world’s best education,
partnering with top universities and organizations to offer courses online.
© 2019 Coursera Inc. All rights reserved.
