Transcript Chapter 8: Data and Knowledge Management

Chapter 11: Business Intelligence and
Knowledge Management
Oz (5th edition)
Data Mining
• Data warehouses are useless without software tools
that process data into information
• Currently Decision Support Systems are called
Business intelligence (BI) software
• BI software takes data and produces information
useful for managerial decision-making
• Data mining refers to the use of tools to extract
information from a data warehouse; business
intelligence is one result of data mining
Data-Mining and Data-Mining Tools
• Data-mining is the process of selecting, exploring, and
modeling large amounts of data to discover previously
unknown relationships that support decision making.
• Traditional data mining tools answer questions about
variables that we think are related
– Query languages (QBE or SQL)
– Report generators
– Multidimensional analysis tools (OLAP or pivot
tables)
– Standard statistical procedures (regression,
ANOVA)
• Knowledge discovery tools are data-mining tools for
finding relationships that are not discernable to the
human eye (see next slide);
Typical Data Mining Tasks Related to
Knowledge Discovery
• Clustering- this activity is designed to take a
population of objects (e.g., customers) and develop
characteristics that can be used to classify them.
You start with no pre-defined classes. Often
clustering is the first step in market segmentation.
• Classification – examines the features of new
objects and assigns them to one of a predetermined
set of classes. Often preceded by clustering.
Clustering could be used to determine
characteristics of customers who respond to
selected types of promotions. Customers in the
same cluster get the same type of promotion
material.
Typical Data Mining Tasks (cont.)
• Affinity grouping (market basket analysis)- This task
is used to determine which things go together.
Typically used to help in cross-selling (e.g., diapers
and beer).
• Prediction – used to determine patterns that can lead
to predictable results. For example, customer churn
or who will default on a loan. Amazon uses items
purchased as gifts to predict the age range of
recipients. Amazon uses your past purchases to
determine what to offer you when you return to the
Amazon site.
Collecting Data for the Warehouse
• Customer loyalty cards have multiple uses, but one
use is to collect data for the data warehouse
• Examples (see textbook for more details)
– Grocery stores
– Web sites
– Harrah’s
– Store related credit cards
• Assurance of a steady flow of data
Multidimensionality or OLAP
• Multidimensional data analysis (or OLAP) enables
users to view data using various dimensions,
measures and time frames (i. e., OLAP)
– dimensions: products, business units,
country, industry (e.g., categories)
– measures: money, unit sales, head count,
variances
– time: daily, weekly, monthly, quarterly, yearly)
• This type of analysis also provides the ability to
view data in different ways (tables, charts, 3-D,
geographically)
• OLAP tools provide for this
• Pivot tables in Excel or Access
Characteristics of OLAP Tools
•
•
•
•
•
•
•
•
•
Primarily used to exploit data warehouses
Provide extremely fast response
View combinations of two dimensions
Enable drilling down (start with broad info and get
more specific)
Produces results as counts or percentages
Conversion of tables to charts/graphs
Usually requires a tailored-made relational database
OLAP applications are widely used by mid-level and
upper level managers
A form of business intelligence software
An OLAP Example
Other Firms that Use OLAP
• Office Depot
• CVS
• Ben & Jerry’s
• DrugStore.com
Customer Relationship Management
(CRM) Systems
• CRM systems are programs to learn more about
customer’s needs and behaviors in order to develop
stronger relationships with them.
• Some sources of data for CRM systems
– Data from Web user’s click stream (see example
about Drugstore.com in the textbook)
– Data from the firm’s data warehouse
– Data from the firm’s customer call centers
– Data from the firm’s help line
– Service and support records
– Customer responses to ad campaigns
–
Goals of CRM Systems
• CRM systems try to use technology to gain insight
into the behavior of customers and the value of
those customers. If CRM works as hoped, a
business can:
– provide better customer service
– make call centers more efficient
– cross sell products more effectively
– help sales staff close deals faster
– simplify marketing and sales processes
– discover new customers
– increase customer revenues
• OLAP and other data mining tools are often available
in CRM software
Summary Thoughts
• CRM software is concerned with data/information
flows between firm and customers
• Datamining is concerned with internal
data/information flows from the data warehouse to
managers (although data originates from external
sources)
• BI software is a more common term for software
once called DSS
• Current BI software focuses on Simon’s intelligence
stage of decision making
• Traditional DSS software focuses more on the
design and choice stages of the Simon model
Summary Thoughts on BI
• Much of BI concerns finding information about
customers
• Datamining and OLAP are often integrated into CRM
systems
• The Web is a popular way to gather BI
• BI on customers promotes targeted marketing rather
than mass marketing
• Third parties often provide BI (e.g., Acxiom and
DoubleClick)
• Overzealous BI efforts are sources of moral and
ethical issues
Executive Dashboards
• A dashboard is a common form of interface between BI tools and
users
– Resembles a car dashboard with clock like indicators and
scales
– Designed so users can quickly grasp business situations
Knowledge Management
• What is knowledge?
– Answer: Knowledge in an organization is the
primarily the collective expertise/experiences of
the organizations employees
• Tacit versus explicit knowledge
– Tacit knowledge is embedded in the human brain
and cannot be expressed easily
– Explicit knowledge is knowledge that exist
outside the brain often in a text format
Examples of Explicit Knowledge
• Written descriptions of best practices for a business
process
• Written knowledge about products, markets, or
customers
• Lessons learned on projects or product development
• Written records of experiences with new approaches
• Examples of successful and failed projects (e.g.,
contracts, proposals, bids, etc)
Knowledge Management
• Capture employee knowledge
• Transfer captured knowledge into a database
• Filter and separate the most relevant knowledge
• Organize the knowledge so that it is accessible to
employees or “push” specific knowledge to
employees based on pre-specified needs
Examples of Knowledge Management
Systems
• Xerox built a Web-based maintenance knowledge
base for field engineers who repair copiers
• AT&T developed a “people finder” database that
provides an on-line directory of “who knows what”
(a knowledge directory)
• HP has a Web-based site that provides knowledge
about competitors, research, products, and
customer satisfaction
• Dow Chemical devised a system to manage its
patents. To keep a patent enforced can cost up to
$250,000. Dow needed to determine which patents
had value. Saved over 40 million in 18 months
Employee Knowledge Networks
• Some tools direct employees to other employees
• Expert can provide non-recorded expertise
• No need to waste money hiring experts in every
department
• Learning from past mistakes saves money
• Employee knowledge network facilitate knowledge
sharing through intranets