Information Systems Development

This essay investigates the current trends in Information Systems Development. It is rare to find an organization in any industry that does not maintain numerous databases to support core business processes. In today's knowledge-base economy, the information compiled and created in an organization is often amongst the company's most valuable assets. Knowledge, its structure, organization and accessibility are commonly acknowledged as key strategic differentiators for many organizations. The development of large-stand alone databases (or data repositories) is no longer the norm for many organizations. Stand-alone databases are being replaced with out-of-the-box database application solutions such as CRM or ERP systems. Legacy stand-alone databases that support key operational processes are being linked to other stand alone or vendor databases in order to support business operations. The development of database systems is rapidly becoming a collaborative effort across enterprises. The true value of organizational information is its ability to inform and empower employees across the company and along processes. In most companies today, databases and the information contained within them are accessed by hundreds or even thousands of employees during the course of the workday. Database development has become the responsibility of every employee who inputs or extracts information from the system in order to support business processes. The issues associated with maintaining database integrity and data stewardship will be discusses in the context of today's shared database development environment.

Information systems have three components: data, processes, and people. The development of information systems requires that users determine what data or information needs to be captured along with associated attributes. The data stored in an information system is only valuable if it provides information about a given process (manual or automated) or activity that is being recorded. The analyses of data and the associated processes are evaluated by a person who completes the "system."

Relational databases provide the data structure for many information systems. Relational databases are made up of a number of tables that have information arranged in rows and columns. The true value of this type of database is that records from one table are related to records in other tables, which allows for easy extraction of related information or records. Prior to the development of the relational database model, data was stored in flat files or computer files that could only be read sequentially. A person reviewing a flat computer file cannot intuitively relate information in a meaningful way. It is important to note that a person is a required element of an information system as opposed to merely the digital information contained within.

Legacy Information Systems

A number of organizations rely on legacy information systems to run core operations and key business processes. Their standalone information database systems were developed in-house to meet production and data management needs. Integrating information systems has become a requirement of today's business world, however. Thus, legacy systems can be viewed as a liability to business process integration and can put a company at strategic risk in the marketplace.

Portfolio Management

Portfolio management of IT assets is a critical task for many IT departments. Compiling an IT portfolio requires a thorough assessment of all applications and systems within an organization and can be a sizable undertaking for an IT department. Portfolio assessment to ascertain business risk involves several steps. Once a list of assets has been compiled, each application should be reviewed for the risk it poses to business operations should it fail. The two biggest risks to business operations are: catastrophic failure of a key system and the constraints placed on an organization to support new business initiatives due to inadequate systems. Knowing the business risks posed by legacy systems allows CIOs and other executive management personnel to make decisions about whether to migrate or replace legacy information systems.

Migration StrategiesCIOs have multiple options for migration strategies, but systems should be migrated selectively and only after a thorough portfolio analysis. CIOs can adopt a strategic approach that leaves core legacy functionality intact and just adds functionality by using newer tools and technologies. Or legacy functionality can be replaced with modern technology by installing packages, replacing the legacy application with an external service, custom-coding a replacement, or a combination of these (Head, 2007).

Vendor Solutions to Information System Development

The implementation of vendor solutions in information systems development has become a common business practice. Vendor solutions require fewer in-house resources for implementation and deployment and allow IT and development staff to work on other strategic solutions. Vendor applications also offer the advantage of interoperability with many other software solutions. Because applications need to mirror business processes from end-to-end, integrated applications are quite desirable.

The biggest factors influencing information systems development are driven by the business requirements of an organization. The following topics are discussed within the business strategy/information system development context:

• Increased user interaction with information systems;
• Examples of information systems;
• Data migration from legacy systems (ETL-extract, transform, load);
• Data governance; and
• Information lifecycle management.

Applications

Explosion of Digital Data

Organizations have been documenting corporate knowledge at an unprecedented rate. There are several reasons for the explosion of digital data being created within organizations large and small. An increasing number of workers have been directly involved with capturing or creating information about business processes, customers, or products. Many organizations have also been converting historical data to digital format and thus have been making the information easily accessible through information systems. Estimates from some industries put the growth of digital information at between 60% and 200% a year in 2007; this includes data in relational databases and unstructured content such as email and network files and other non-relational databases (Enterprise Content Management, 2007).

Organizations have responded to the explosion of digital data by creating a new generation of information systems that allow for better storage and retrieval of knowledge-based assets. Enterprise data warehousing (data marts), management information systems (MIS), and content management systems (CMS) are a few of the information systems that have been deployed in organizations.

Data Warehousing

An enterprise data warehouse is a large database that is meant to store a company's historical data and corporate knowledge. The data warehouse is an information database that may be surrounded and accessed by any number of enterprise systems including: a customer relationship management system (CRM), supply chain management system (SCM), or corporate performance management system. Data warehouses have been popular because they collect vital company information in one central information system and can eliminate the need for multiple information database systems.

Many first generation data warehouses have not lacked in raw data, and many have done a good job of supporting enterprise systems that are highly transactional in nature. For instance, these systems have done a good job of linking together financial transactions (accounting records) with their associated operational transactions (purchase orders, deliveries, inventor movements, invoices, etc.), creating a joined path back to the data points needed to determine the actual cost of acquiring a product from a supplier (Foulkrod, 2007). Where most data warehouses have not done such a good job is in aligning financials to business operations on a granular level. This "content deficiency" has been caused by a lack of ability to tie a business context to much of the data in the warehouse (Foulkrod, 2007). Business rules need to be associated with data in the warehouse -- the process of applying context to enterprise data can be described as tailoring or content enrichment. In other words, a person must interact with the raw data and apply design principles, with the ultimate objective being the extraction of information that will support business decisions and not just transactional processes from the warehouse.

Dimensional Database Model vs. Relational Model

Developers have designed data warehouses, in part, in what is known as a dimensional model. This model is designed with the end-user in mind and provides a simple and understandable way to extract information about a particular business process -- which is generally the focus for most users. Dimensional modeling is based on meeting business needs while normalization (3NF) (relational databases) is based on data modeling. A dimensional database contains multiple business processes and can best be envisioned as a data cube with 3 or 4 dimensions. Users are able to access a slice along any of the dimensions and therefore identify a very specific data point at the intersection of any of the dimensions.

Most data warehouse (DW) environments need data stored in both 3NF (relational) and dimensional models, not either/or. Relational modeling supports business intelligence reports and the analysis of data. Normalized design supports history, changing dimensions, data integrity, and quality. Accordingly, under such conditions, a normalized model is the best strategy in this sort of situation. This varied environment is actually a DW using a normalized structure to archive data and manage change data capture (CDC) with a data mart in a dimensional model to enable BI reporting and examination. Both of these two approaches can be utilized in the same database using different logical areas, such as a schema in an Oracle environment.

Relational databases are relatively simple to design and operate and are perfect for storing large volumes of data with high transaction rates. The drawback for the end-user extracting information from a relational database is two-fold:

• The information may not support analysis from a business standpoint; and
• Typical queries are difficult to design and run and require a trained database manager to run.

Dimensional databases may require more input from users and teams to design the business parameters around the tables and associated dimensions. Benefits of the dimensional model are:

• Data results are more intuitive for users in extraction and interpretation;
• Dimensional databases are easy to maintain because information is stored in the same way it is viewed; and
• Dimensional databases out-perform relational databases in returning results (faster delivery of queries).

Management Information Systems

Any discussion of information systems development needs to focus on what information organizations, executives, and boards expect to get from an information system. A management information system (MIS) is one of the most popular management database tools in business. An MIS is capable of collecting and reporting key performance indicators and can replace the basic P & L -- based performance measurements with data that is much more meaningful in the overall context of business operations (Fitzpatrick, 2007).

The design of an MIS at Mercury Asset Management is an example of a MIS with a very dimensional feel in its design. Essentially, this was a 3D virtual grid used to examine a product's success at winning customers in each source market. To do this, it compared a range of data including fund sizes, redemptions, growth, accounting costs, and revenues. With this type of matrix, a firm might find that although it sold more of a certain fund in a particular market, this was not the most profitable fund in that market due to the costs associated with, for example, accounting. A single P&L figure can miss the story. As stated by Magnus Spence, partner and chief operating officer at Dalton Strategic Partnership, "The key to MIS is that it's not P&L accounting" (as quoted in Fitzpatrick, 2007).

Requirements of an MIS

One can picture a management information system as an application sitting on top of the enterprise data warehouse (ideally). The subset of data that is needed to populate the MIS can be very specific to a business unit (e.g, sales, marketing) and supports analytics rather than transactional processing. Top requirements of an MIS, in this example from a professional services organization, are (Fitzpatrick, 2007):

• Client satisfaction/relationship retention,
• Key client analysis,
• Productivity/operational effectiveness,
• Performance-linked rewards,
• Costs and profitability, and
• Daily sales reports, including revenues and margins.

Content Management Databases for Knowledge Workers

Content management systems have been popular at organizations for many years. First-generation content management systems stored documents in simple standalone databases that allowed users to query for documents by title, keyword, or other metadata tags. Content management systems often had version controls and a workflow component for routing, approval, etc.

Contemporary content management systems are typically server- and web-based and are designed to house the unstructured data that is created within an organization (estimated at 80% of generated content). "Content management represents a critical technology to help address this explosive growth. As organizations consider the role of content management in the broader information management context, they are choosing to standardize on an ECM platform in order to build an information infrastructure across the enterprise. A common infrastructure reduces the cost of developing and deploying line-of-business solutions, improves integration with business applications, and enables the proactive management of the information lifecycle" (Enterprise Content Management, 2007).

It is important to reiterate that in many organizations, knowledge is the product or service that is offered to customers. Knowledge is generated at every level of the organization and also needs to be at the fingertips of every worker who interacts with a business process. Information systems need to be able to accept large volumes of unstructured data as well as search for relevant information across enterprise applications. Collaborative and user-maintained information databases are vastly different from the large transactional information systems that were once the norm at many companies.

Issues

IT departments have long played a role in the development, administration, and maintenance of information systems within companies. The topics covered so far in this essay reveal that the nature of corporate information systems has been rapidly changing, and the impact on IT staffs has been acute.

Continual Challenges to IT Organizations

"The continuous growth of information presents a challenge to IT organizations and, if it remains unchecked, will become a crisis that impairs the ability of IT to meet the strategic and business needs of the organization. Information, including structured data (typically stored in relational databases) and unstructured content (stored in file systems, content management systems, email servers, and more), is growing between 60 and 200 percent per year and is generating a number of negative side effects: storage costs draining IT budgets, eDiscovery costs paralleling the explosion in growth, and employee productivity limited by time wasted searching for information that is often never found and ends up being recreated" (Enterprise Content Management, 2007).

Role of the Database Administrator

With the rise of enterprise-wide data storage, the role for the database administrator has evolved. "The role of the DBA has changed considerably over the years. Today, DBAs play a strategic role in the overall success of business and often work directly with an organization's leaders to ensure that overall operations are successfully driving the company forward. It used to be that the DBA was the back-office person who understood the scheme of the database and took direction from the application developer. And the application developers were the people that dealt with the end users and customers" (McLean, 2006).

The ETL Function

As previously discussed, the information in company information systems is sought after by every employee who is interacting with a business process. Database administrators are responsible for insuring that information is secure and accessible. The rise in popularity of enterprise-wide applications and data repositories has put a focus on a database administrator function called ETL (extract, transform, load). ETL is the process of populating an enterprise data warehouse from existing standalone databases.

Just how big is the scope of ETL work? "Ahead of simplifying and integrating their HR systems, Novell's information support services team identified 19 different applications just to serve their hiring process. Shell Oil Co. had 85 ERP systems, with each implementation customized to fit a specific business requirement" (Ananthakrishnan, 2003). Each of these applications had its own underlying information database repository that needed to be processed before the data could be migrated to an enterprise system.

One database administrator described this function in the following way. "In my particular function, the work entails the batch-loading of data from external sources into the databases and making several types of exports to send updates to various parts of our online or distribution systems" (McLean, 2006).

Database administrators have taken on the integral role of processing raw data and corporate information and repackaging it in a transformed state to enterprise-wide information systems. The following steps have been followed to insure that the best possible raw data goes into the new information database (McGilvray, 2006):

• Cleanse: Review records from various source databases and clean up records.
• Transform: Transformation programs take source data and create destination data -- this step insures data meets business needs.
• Prepare: Standardize data formats.
• Migrate: Load data into a destination database (data warehouse).

Data Governance

A recurring theme within business applications and systems deals with the implementation of documented standards to insure process quality. The costs associated with migrating terabytes of data from system to system is staggering and has sapped IT and development resources in many organizations. Data governance, defined as the management of the availability, usability, integrity, and security of the data, is a top priority for organizations when developing information systems.

As organizations implement enterprise-wide applications and databases, there are many issues surrounding the migration of data to new systems. Organizations have been generating volumes of data for many years with much of this legacy information stored in proprietary databases that served production systems or other business units such as finance or sales. There are a number of challenges associated with tying together data and information from disconnected systems within an enterprise and can include (McGilvray, 2006):

• Redundant information stored in separate databases (sales and marketing may have the same customer information in two separate databases);
• No common field way to link data from separate database systems -- unique or key fields are a requirement for normalized databases; and
• No dedicated team assigned to evaluate and integrate data into the enterprise system.

The data governance team should include representatives from all business units within an organization and can impact information systems development in several ways. Some of the responsibilities of the data governance team are to:

• Set up business rules for the ETL (extract, transform, load); process.
• Define and document all rules associated with data integrity for data that will be loaded into the data warehouse system;
• Insure ongoing quality of data in the system;
• Define roles and responsibilities of data stewards and communicate policies to the organization; and
• Show how data changes impact business and affect other areas in the organization.

Information Lifecycle Management

The management of information over its lifecycle has long been the responsibility of librarians and records management professionals. Today, the lifecycle of digital data has been redefined and expanded as the scope of data that needs to be retained has increased.

The information lifecycle is (generally) composed of the following steps:

• Creation of document (by human, process, etc.);
• Acquisition (incorporation of item -- physically or virtually -- into collection);
• Cataloging and identification (capture of metadata -- title, author, origination date, file type, etc.);
• Storage (physical storage or digital storage);
• Preservation and access (active or archived status -- allows users to find information); and
• Compliance and control (implemented due to federal mandates for privacy and security).

Compliance & Control

Lifecycle management of information is a joint effort between IT and the business unit or units responsible for managing content within an organization. "Compliance is a requirement for the organization, but can be a burden on the line of business. Content management systems, which inherently bring content under greater control, can help fulfill compliance mandates by identifying which pieces of information must be controlled or retained and by enforcing those policies. IT also plays a part by providing the appropriate service levels to ensure there isn't a technical break-down that would cause the organization to fall out of compliance. Pairing the content management system with ILM practices results in a policy-driven IT architecture that ensures the right information is preserved and allows an organization to respond quickly and confidently to regulatory or legal scrutiny" (Enterprise Content Management, 2007).

"How IT handles a piece of information can often be described in 'service levels' such as response time, recovery time, retention periods, and risk of data loss. High service levels can guarantee the business has access to the information it needs, but can be very costly; hence IT's desire to use ILM practices that reduce their overall costs" (Enterprise Content Management, 2007). It is critical that a business have reliable access to information that protects the organization from litigation or breach of compliance.

Conclusion

This essay has investigated some of the trends in information systems development within organizations. Topics include the development of enterprise-wide information systems rather than standalone databases. Business initiatives have driven the development of corporate information systems as the knowledge and data associated with business processes has become critical to operational success. Data warehouses that store comprehensive corporate knowledge and historical data will remain a focus of database development in large organizations. The information stored in data warehouses will undergo transformative processes as the data becomes more closely associated with key business processes. IT staff and cross-functional data governance teams will oversee the contents of data warehouses as stewards of data quality, relevancy, and integrity. IT staff will need to manipulate, secure, track, and archive ever increasing volumes of digital data. Compliance mandates will require the implementation of information lifecycle management policies that insure that most of the digital data created will be around for a long time.

Terms & Concepts

Dashboard: A digital dashboard, also known as an enterprise dashboard or executive dashboard, is a business management device used to visually determine the status (or "health") of a business enterprise through identifying key business indicators.

Data Governance: Data governance (DG) refers to the management of the overall availability, usability, integrity, and security of the data utilized by an enterprise.

Data Warehouse: Where the majority of an organization's historical data is maintained. It could be said to be an organization's 'corporate memory'.

Extract, Transform, Load (ETL): The process of loading data into a data warehouse. This process involves extracting data from one or more outside source, transforming (manipulating) data to fit a business need, and loading the data into a data warehouse.

Information Systems: The persons, data, records, and activities that process data and information -- both manual and automated processes.

Information Lifecycle Management (ILM): The technique of administering specific policies to achieve the effective management of information throughout its useful life.

IT Asset Portfolio: The inventory of all an organization's IT applications and systems for inventory and risk assessment purposes.

Management Information Systems (MIS): Information systems that are used to analyze information systems that record and report operational activities in the organization.

Metadata: Data about data. Metadata describes the data. For example: title, publish date, author, or URL.

Relational Databases: Database model in which data is stored in the form of tables and the relationship among the data are also stored in the form of tables.

Service-Level Agreement: A contract or predefined agreement that outlines key deliverables to a customer or in the IT department to a business unit.

Structured Data: Data found in a relational database that has defined structure and attributes.

Undefined Data: User-generated content typically found in file systems, e-mail, or content management systems.

Bibliography

Ananthakrishnan, V. (2003). The systems complexity quandary. Siliconindia, 7(9), 44-45. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=11118672&site=ehost-live

Fitzpatrick, N. (2007). Don't pay the boss's airfare. Global Investor, (199), 23-24. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&A N=24330825&site=ehost-live

Foulkrod, J. (2007). The data warehouse content gap. DM Review, 17(6), 16-18. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&A N=25302368&site=ehost-live

Geiger, J. (2007). Ensuring quality data. DM Review, 17(1), 43-44. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=24656869 &site=ehost-live

Gregory, R., Beck, R., & Keil, M. (2013). Control balancing in information systems development offshoring projects. MIS Quarterly, 37 (4), 1211 -- A4. Retrieved November 26, 2013 from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=91906210&site=ehost-live

Head, B. (2007, June 5). Managing the legacy portfolio. CIO, 3-3. Retrieved August 9, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=25377113 &site=ehost-live

IBM Corporation (2005). Concepts of dimensional data modeling. Retrieved Aug 14, 2007, from IBM Informix Database Design and Implementation Guide Web site: http://publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp?topic=/com.ibm.ddi.doc/ddi222.htm

Kamath, J. (2007, June 12). CIOs most prompt board rethink on IT, says analyst. Computer Weekly, 8. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=25636457&site=ehost-live

Kawamoto, T., & Mathers, B. (2007). Key success factors for a performance dashboard. DM Review, 17(7), 20-21. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=25733752&site=ehost-live

Long, J. (2007, June 26). Measure IT alongside business aims. Computer Weekly, 20-22. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&A N=25754190&site=ehost-live

McGilvray, D. (2006). Data governance: A necessity in an integrated information world. DM Review, 16(12), 24-26. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=23795643&site=ehost-live

McLean, C. (2006). Database administrators: Multitasking for advancement. Certification Magazine, 8(1), 30-40. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=19364593&site=ehost-live

Narayanaswamy, R., Grover, V., & Henry, R. M. (2013). The impact of influence tactics in information system development projects: A control-loss perspective. Journal of Management Information Systems, 30 (1), 191-226. Retrieved November 26, 2013 from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=89799096&site=ehost-live

Sherman, R. (2007). The trial-and-error method for data architecture. DM Review, 17(1), 39. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=24656865&site=bsi-live

Srivastava, S. C., & Teo, T. H. (2012). Contract performance in offshore systems development: Role of control mechanisms. Journal of Management Information Systems, 29 (1), 115-158. Retrieved November 26, 2013 from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=7962 9588&site=ehost-live

Theobald, J. (2007). Strategies for testing data warehouse applications. DM Review, 17(6), 20-41. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=25302369&site=ehost-live

Suggested Reading

Alaa, G., & Fitzgerald, G. (2013). Re-conceptualizing agile information systems development using complex adaptive systems theory. Emergence: Complexity & Organization, 15 (3), 1-23. Retrieved November 26, 2013 from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=906 19661&site=ehost-live

Beasty, C. (2007). Taking out the trash. CRM Magazine, 11(3), 41. Retrieved August 9, 2007, from EBSCO Online Database Academic Search Premier. http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=24287385 &site=ehost-live

Mishra, A., & Mishra, D. (2013). Applications of stakeholder theory in information systems and technology. Engineering Economics, 24 (3), 254-266. Retrieved November 26, 2013 from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=89252458&site=eh ost-live

Reed, D. (2007). Who controls your data? Precision Marketing, 19(15), 27-29. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=25379484&site=ehost-live

Shrivastava, A. (2007). Information explosion creating opportunities for storage professionals. Certification Magazine, 9(8), 26-29. Retrieved August 9, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=257 33023&site=ehost-live