Experiences in Strategic Information Systems Planning

...

(1) increased ability to share information to address subunit interdependence, (2) reduced ability to meet unique subunit information requirements, and (3) changes in the costs of information system design and implementation.

The first factor may have two different impacts on an organization: (1) improved managerial information for organization-wide communication and (2) operational coordination between interdependent parts of the organization.

The second factor might reduce local flexibility by data integration in two ways: compromise and bureaucratic delay.

The third factor deals with the impact of data integration on design and Implementation costs depends heavily on the situation and why and under what circumstances up-front costs are higher, and why and under what circumstances long-term costs are lower.

The Author continues to explain why the rational model is only a part of the picture and there are many other factors that play a role in data integration. Power and Politics is one such segment which explains that the possibility of changes in the power balance can cause resistance to a data integration effort by those concerned they might lose out, regardless of arguments taking a total organizational benefit point of view. The question of who gets the benefits and who pays the costs is not an idle one.

Finally the Author concludes in general it will not be cost-effective to integrate all of an organization's data. Greater interdependence and greater reliance on computer-based information and changes in the organizational climate will probably shift the balance toward the need for greater (but not total) data integration in many firms, heightening the practical and academic importance of this area.

Article 3:

Managing the Data Resources: A Contingency Perspective

The article explains about the increasing emphasis the corporations are placing on data management. After analyzing 31 case studies with 20 different firms on managing data resource it was found that there is no single dominant approach to improve management of data but rather adopt diverse multiple approaches which differ in business objective ,organizational scope, planning method and product. The problem of unmanaged data is quite real and exists in a broad range of organizations.

A summary of current approaches and research approach is described for better management of data. Current approaches are categorized into 3 different types: 1)Approaches with a technical focus 2)Approaches with a focus on organizational responsibilities and 3)Approaches with the focus on top-down business related planning. The third approach has received a great deal of attention due to its success in providing data to support business needs. The Research approach consisted of exploratory case studies conducted across multiple firms and industries. An assumption, supported by the findings, was that the similarities in data management faced by large organizations outweigh differences among industry groups.

Contingency approach to data management consisted of a framework which represented 4 key elements: 1) The Identification of a business objective 2) The scope of the data management project 3) The data planning method.4) The “Product” of data management effort.

The 5 data management products found are 1)subject area database for operational systems 2)Common systems 3)Information databases 4)Data access services and 5)Architectural foundations for the future.

The 4 data planning methods established were 1) strategic data planning 2) targeting high impact areas 3) 80/20 planning methods and 4) No planning process.

In the bounded scope of the data management method is to carefully select the scope.

It was observed that there were organizational issues affecting data management implementation. The issues are as follows: short term and long term tradeoffs in resource allocation, the centralizing tendency of data management, Impact on the IS culture, New responsibilities for user management and the process of effectively introducing innovations into the organization.

Author concludes that there is no single clear-cut approach to data management but a wide range of options exist to fit the needs of a particular business.

Article 4:

A Model of Data Warehousing Process Maturity

Arun Sen, K. (Ram) Ramamurthy, and Atish P. Sinha

The article is regarding the argument that the maturity of a data warehousing process (DWP) could significantly mitigate such large scale failures and ensure the delivery of consistent, high quality, data in a timely manner. In light of the critical importance of data as a corporate resource, the need for a maturity model for DWP has raised. In this paper is described the design and development of a five-level DWP maturity model (DWP-M) over a period of three years. A unique aspect of this model is that it covers processes in both data warehouse development and operations. The final model was evaluated by a panel of experts; the results strongly validate the functionality, productivity, and usability of the model. The initial and final DWP-M model versions, along with illustrations of several key process areas at different levels of maturity has been presented in this article.

Just as CMM has been useful in reducing defects in a software development process, a mature DWP addresses many issues surrounding the development and management of a data warehouse. In addition to addressing data quality concerns, a mature DWP can be expected to provide organization to define and deliver projects with predictable durations.

DWP maturity revolves around data quality management, ETL design, metadata management, data change management, data warehouse governance, end-user cube design, etc.—activities that do not fall under the purview of traditional software development. The main contribution of our work is in designing a DWP maturity model by identifying, defining, and accommodating those aspects that pertain specifically to DW. The paper is organized as follows: Section 2 provides the motivation for developing the DWP maturity (DWP-M) model. Section 3 reviews the extant literature on maturity models. Section 4 first presents the framework that has been recently proposed for conducting design-science research in IS, and then describes how we employed this framework to design, develop, and evaluate the DWP-M model. Section 5 discusses the contributions of our study