Page 52 - CITP Review

In recent years, entities are capturing and storing more and more data. The key is to identify the life span
of data being collected and to provide the appropriate archive system, storage, and management for
permanent data. A data warehouse is often employed to archive permanent and semipermanent data
needed for BI or other strategic purposes.

The entity would need to use archival techniques for developing its policy for archiving various types of
information and data. The information or data could be categorized by type where useful lives are
consistent within each category.
For instance, email might consistently have a useful life of 18 months and be archived accordingly.
Attachments to email, however, might need to be archived based on the type of document it is, and a
different rule applied to the attachment.

Similarly, legal documents and work product have special requirements that might require archiving for
several years. The Securities and Exchange Commission (SEC) requires data and documents related to
audits and reviews of financial statements to be kept for seven years. Informal meeting notes in
electronic documents are likely to have short lives.

Destruction
Data that reaches the end of its life span should be recognized as such, and provisions should be in place
to appropriately destroy that data.

Different types of data have different archival lives, and some of the life spans differ between entities.
Data destruction policies need to be developed based on contractual, legal, and other constraints.

Data should not be stored infinitely without this consideration. To do so incurs costs and risks, including
legal risks. Each entity should work with legal counsel concerning specific criteria for the archival and
data destruction policy.

Infrastructures and platforms

The processes described in data consolidation, data cleaning, data transformation, and data reduction
are the steps necessary for data preparation that leads to effective data mining and BI. It is estimated
2
that 80% of the effort to build an effective BI or data mining system is expended in data preparation;
however, before data preparation begins, there needs to be an understanding of the infrastructure.

Types of infrastructure and platforms typically employed
The choice of infrastructure is important in developing a database system for data analysis and
reporting. Sometimes an existing platform is suitable, and sometimes the entity has to develop an
effective solution to integrate the data warehouse, sources of data, and analytical and reporting tools.

2
Ramesh Sharda, Jay E. Aronson, Efraim Turban, and David King, Business Intelligence: A Managerial Approach,
2nd ed. (Boston: Pearson, 2011), 150.

© 2019 Association of International Certified Professional Accountants. All rights reserved. 2-6