How to Equip Students with Methodologies and Tools for Capturing

Rapidly Changing Environments through Computer

Supported Education

Ushio Sumita and Jun Yoshii

Graduate School of Systems and Information Engineering, University of Tsukuba,

1-1-1 Tennoudai, Tsukuba, Ibaraki, 305-8573, Japan

Keywords: Matured Market Economy, Customer Segmentation, The 70-30 Principle, Intermediary DB, Profile Vectors.

Abstract: In the midst of the global mega-competition, the most competitive battle fields of economy, often referred to

as the tripod consisting of the United States, EU and Japan, has entered the matured market economy where

consumers are interested in acquiring goods and services to fit their particular needs, demanding a variety of

products and services in small quantities. Accordingly, corporations now have to deal with segmented

submarkets which change quite rapidly. For achieving the leading competitiveness in this new environment,

it is no longer sufficient to capture the entire market as a whole. Instead, it is necessary to trace and analyze

the segmented submarkets separately. In order to overcome this difficulty, methodologies and tools are

needed for extracting effective managerial implications from the massive data collected through the Internet

with speed and accuracy. The purpose of this paper is to demonstrate how to equip students with such

methodologies and tools through computer supported education.

1 INTRODUCTION

The growing market economy may be characterized

by the fact that consumers share the sense of lacking

goods and services for consumption and are eager to

possess what others have. In contrast, in the matured

market economy, consumers tend to pursue

individual tastes in consumption so as to maximize

their own sense of satisfaction. In other words,

consumers are interested in acquiring goods and

services that others may not have but fit their

particular needs. Naturally, this trend results in a

variety of products and services in small quantities

and the market segmentation becomes extremely

important. A typical successful R&D strategy in this

stage would be the market-in strategy, where a

variety of products are introduced into the market in

small quantities in response to particular needs in

particular market segments. Such products in the

matured market economy would have much shorter

life cycles than those in the growing market

economy, causing rapid changes in the segmented

submarkets.

As long as the real economy is concerned, the

economies of scale is always present. Since the

matured market economy requires more detailed

marketing strategies for individual segmented

submarkets, the efficiency resulting from the

economies of scale tends to diminish. In other words,

if corporations have to deal with separate segmented

submarkets in a one-on-one manner, the profit

margins would inevitably decrease. In order to

overcome this difficulty, methodologies and tools

are needed for extracting effective managerial

implications from the massive data collected through

the Internet.

The central approach for achieving this goal

would be to apply the 70-30 principle, proposed by

the authors in (Sumita and Yoshii, 2012), to the

information processing procedures, where such

procedures for separate segmented submarkets are

designed 70% in common with remaining 30% for

customization so as to cater for peculiarities of

individual submarkets. This observation would be

valid across many different industrial sectors.

Accordingly, from a pedagogical point of view, it is

very important to familierize students with the 70-30

principle applied to information processing, no

matter what industrial segment they plan to enter

after graduation.

493

Sumita U. and Yoshii J..

How to Equip Students with Methodologies and Tools for Capturing Rapidly Changing Environments through Computer Supported Education.

DOI: 10.5220/0004367904930496

In Proceedings of the 5th International Conference on Computer Supported Education (CSEDU-2013), pages 493-496

ISBN: 978-989-8565-53-2

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

The purpose of this paper is to demonstrate how to

equip students with methodologies and tools for

implementing the 70-30 principle in information

processing through computer supported education.

In this paper, we focus on dynamic customer

segmentation in e-marketing for enhancing CRM

(Customer Relationship Management). A

pedagogical training scheme is illustrated explicitly

through computer supported education. Furthermore,

we clearly outline the procedures for extracting

managerial implications from the massive data

collected through the Internet.

2 DYNAMIC CUSTOMER

SEGMENTATION IN

E-MARKETING FOR

ENHANCING CRM

More than a decade has passed since the Internet

gained its significant presence in the world. It has

penetrated into many aspects of business practices

and has been drastically changing the traditional

business models in almost every industry. In the

retail chain business, for example, it is now possible

to collect and accumulate massive data from the

market via a POS (Point of Sales) system and utilize

them so as to develop effective marketing strategies

for enhancing sales of products. An extensive

literature exists for analyzing consumer purchasing

behaviors based on POS data, represented by

(Taguchi, 2010; Eugene, 1997; Ishigaki et al., 2011;

Yada et al., 2006) to name only a few.

The problem we face here is the excessive

computational burden, where the tremendous

amount of POS data collected from the market has to

be analyzed repeatedly in a timely manner. In order

to overcome this difficulty, we introduce the concept

of profile vectors as an intermediary information

base between various analytical engines and the DB

of POS data.

Figure 1 depicts the basic framework for

implementing dynamic customer segmentation. Here,

a variety of profile vectors, such as CPV (Customer

Profile Vector), PPV (Product Profile Vector) and

SPV (Store Profile Vector), are automatically

constructed and updated periodically from the DB.

These profile vectors are then used by different

analytical engines, producing the standard reports

Figure 1: Dynamic customer segmentation for enhancing CRM.

CSEDU2013-5thInternationalConferenceonComputerSupportedEducation

494

from the basic analysis as well as some ad hock

reports derived from characteristic analyses

specified by a manager through the graphic interface.

Furthermore, these results are used to update

customer segments dynamically so as to yield

different marketing strategies applied to different

customer segments.

CRM typically means that the lifetime value of a

customer is to be maximized by maintaining two

way communications between the customer and the

company through the Internet. This concept is

limited in that the potential customers are not

addressed explicitly. By combining POS data with

transaction data on the Internet, not necessarily

linked to purchasing, it is now possible to capture

the entire market as depicted in Figure 1, where the

market is decomposed into 9 segments: (Existing-

Active, Existing-Sleeping, Potential) × (Not

Important, Normal, Important). The arrows (1)

through (7) indicate the desirable changes of the

market for the company, whereas the arrows (8)

through (12) represent the changes of the market to

be avoided. The new marketing approach for

enhancing CRM would then be to devise strategic

moves so as to promote the moves along favorable

arrows and prevent the moves along unfavorable

arrows. Since such customer segments have to be

updated dynamically, the profile vector approach

becomes crucial for containing the underlying

computational burden. This example demonstrates

the importance of the 70-30 principle in e-Marketing.

For establishing a base for computer supported

education through this example, where students can

learn how to develop and maintain the dynamic

customer segmentation system illustrated in Figure 1

based on the 70-30 principle, a computer simulator

for the dynamic customer segmentation system is

installed in a server at Sumita Research Laboratory

in parallel with the system developed at the

collaborating e-business company through the joint

research project. Real data collected from the

Internet are fed into the simulator once a week. This

simulator enables students to actively get involved in

the decision process for development and analysis of

e-marketing strategies.

3 CONCLUSIONS

In the matured market economy, consumers are

interested in acquiring goods and services that others

may not have but fit their particular needs so as to

maximize their own sense of satisfaction. Naturally,

this trend results in a variety of products and

Figure 2: Customer profile vector.

Figure 3: Product profile vector.

services in small quantities and the market

segmentation becomes extremely important. Such

products introduced in response to segmented

submarkets would have much shorter life cycles,

causing rapid changes in the segmented submarkets.

In this paper, we propose a general scheme

involving methodologies and tools for capturing

such rapidly changing environments.

The proposed general scheme is based on the 70-

30 principle, proposed by the authors in (Sumita and

Yoshii, 2012), applied to the information processing

procedures, where such procedures for separate

segmented submarkets are designed 70% in common

with remaining 30% for customization so as to cater

for peculiarities of individual submarkets. The key

success factor for development of the general

scheme is to introduce profile vectors as an

CPV (Customer Profile Vector)

Part I

Customer ID

Sex

Age

…

Pa rt II

Time Period X(1) [ Tota l Quantity,

Total Monetary Amount,

Tota l Qua ntity of Ca tegory (1),

…

Tota l Qua ntity of Ca tegory (K),

Tota l Monetary Am ount of Ca tegory (1),

…

Tota l Monetary Am ount of Ca tegory (K),

… ]

Time Period X(2) [ Tota l Quantity,

Total Monetary Amount,

Tota l Qua ntity of Ca tegory (1),

…

Tota l Qua ntity of Ca tegory (K),

Tota l Monetary Am ount of Ca tegory (1),

…

Tota l Monetary Am ount of Ca tegory (K),

… ]

…

PPV (Product Profile Vector)

Part I

Product ID

Product Name