Managing Information Technology

Chapter 4 • The Data Resource 107

3.Acquire and Maintain Once the best sources for
data are identified and selected, data capture systems
must be built to acquire and maintain these data.
Changes in data need to be broadcast (via master data
management techniques and technologies) to all
databases that store these data. Users of the data need
to know when the data are refreshed and perhaps
automatically be informed of exceptional conditions
(e.g., inventory being out of stock, stock price below
a critical level, or receipt of an especially large
customer order). Appropriate applications systems
need to be built to track data acquisition and transfer.
For example, suppose electronic files of customer list
data are sent to telemarketing vendors for a promo-
tional campaign and results are returned via the
Internet. A system is needed to confirm that all files
were sent and received, that all customers on the list
were called, and that a status is received on each.

4.Define/Describe and Inventory A basic step in man-
aging any resource is defining what is being managed.
For a real estate manager, each property must be
described, standards and scales must be set to define
the size and shape of each building or land parcel, and
terminology must be defined to refer to different
pieces of each building. Similarly, in managing data,
each data entity, data element, and relationship must be
defined, a format for storage and reporting established,
and the organization of the data described so users
know how to access the data. As mentioned earlier, a
metadata inventory catalog must be maintained,
usually using a DD/D, where all data definitions and
descriptions are kept, volume statistics on data are
maintained, and other data about data (such as access

rights and integrity rules) are stored. All users can go

to the metadata repository and data dictionary to find

out what data exist and what the data mean.

5.Organize and Make Accessible Databases need to

be designed so that data can be retrieved and reported

efficiently and in the format that business managers

require. Data should be arranged and stored so that

information can be produced easily. Although most

of the work here is rather technical, this physical

arrangement of data cannot be done unless potential

uses of the data are well defined, and this task is best

done by business managers. The two aspects of data

usage necessary for proper organization are what

data are required and how the data are to be selected.

For example, database designers need to know if

customer data will be selected by markets, by

geographical regions, by what products they have

bought, through what sales staff they buy, or by

other criteria. Orders of magnitude improvements in

processing speed can be achieved when the data

organization is well-tuned to the processing

requirements. Of course, wise choices of database

designs can similarly achieve significant reductions

in the cost of maintaining and processing data.

One highly popular method for making data

accessible to many people in an organization for

decision making and business intelligence is the data

warehouse (see Chapter 5 for more on data warehous-

ing). Figure 4.7 depicts how a large division of a

furniture manufacturer recently implemented a data

warehouse. Prior to the creation of the data warehouse,

the company operated several legacy applications

systems, each containing data difficult to extract but

Data

Capturing

and

Processing

New

Applications

Systems

Operational

Store

Data

Warehouse

Data

Analysis and

Presentation

Single Enterprise

Data Repository;

Organized for

Analysis

Data Cleansing

and Transforming;

Security Issues

Legacy

Applications

Systems

Analytical

Tools

Reporting

Tools

Actions from Decisions

FIGURE 4.7 The Data Warehouse