Strategies for Data Management Leadership: Why, Who, What, and How Copyright CAUSE 1994. This paper was presented at the 1993 CAUSE Annual Conference held in San Diego, California, December 7-10, and is part of the conference proceedings published by CAUSE. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, that the CAUSE copyright notice and the title and authors of the publication and its date appear, and that notice is given that copying is by permission of CAUSE, the association for managing and using information technology in higher education. To copy or disseminate otherwise, or to republish in any form, requires written permission from CAUSE. For further information: CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301; 303449-4430; e-mail info@cause.colorado.edu Strategies for Data Management Leadership: Why, Who, What, and How Lore A. Balkan, Gerry W. McLaughlin, and Richard D. Howard The critical issue is not one of tools and systems but involvement in the quality efforts of the business units.[1] Overview A recent advertising campaign for IBM calls for "new ideas for new challenges". This has three elements of relevant to those of us who work with information. First there is a strong theme of change. New is the norm and change is the standard. The second theme is that of survival in the face of challenges. The way we do everything is subject to question because our world has changed and promises to change at an even greater rate in the future. This theme acknowledges the ever present threat that if we fail to adjust, we and our services will become obsolete. The third theme is less apparent, but provides the keys to open new doors as we close the old. It is quite simply one of creating new vision from ideas generated at all levels of our organizations. Unlike days gone by when demonstrated knowledge and skill were required before ideas were solicited, today we ask the people at the pulse of every activity to interact and generate innovative ideas. The necessity and the ability to acquire new knowledge and skills is accepted as the rule. The road to success is paved with ideas. These ideas are then molded into plans and resources are reallocated to support training on the new and improved practices and products. The key challenge for organizations is to make better decisions and provide better support for their stakeholders. Information support must enable: 1) recognition of the need to make a decision, 2) identification of alternatives, 3) a call to act, and 4) the ability to verify and defend the action. The successful use of information to this end depends on an information support structure that assures the quality and availability of relevant data and that restructures the data for the decision makers. Not surprisingly, the old ways of doing information support cannot support our new ways of doing business. The disappearance of the middle management of the downsized organization, the advent of intelligent devices, and the refinement of strategic management, all are the components of the new challenge. To meet this new challenge, we need new ideas about providing information support for the entirety of our organizations. Current Culture The last several years have seen rapid changes in the technology with which we manage our organizational data. Complementary changes now need to occur in our organizations and in our concepts about information systems. We are witnessing a simultaneous push by technology and pull by management to deliver useful information. It is naive to assume that all of the required management information in fact resides in our historical legacy systems, and that the challenge is to simply implement new technology to deliver data on demand to an expanding clientele. What is too often overlooked is that management's requirements are driven by new needs to perform analyses related to both longterm and shortterm decisions and then to evaluate the results of those decisions. The operational systems that perform the daytoday transactions have not been designed to support this activity and quite likely do not contain sufficiently standardized or normalized data for the integrated, recombined, and longitudinal views required for analysis. Existing internal and external reporting functions often use informal procedures to obtain data and to interpret the variables. There is not a predominance of historical files, or census files, or standardized data. Furthermore, there is often no formal assignment of responsibility for data management, nor policies governing the process by which those involved with the operational systems capture, store, define, secure, or provide the data to management or do reports for distributed decision makers. Typically when a management call for information fans out to a variety of operational areas each responds by providing data from the perspective of its functional activity. The result is that 1) the executive will be drowning in data with no options to apply analysis and transform the data to useful information, 2) the executive will receive multiple, biased, and conflicting informational answers from a variety of functional areas, 3) the information provided will be based on incomplete assumptions about the desired analysis and incomplete data in terms of integration with other functional areas, and 4) the failure to properly obtain the information will produce organizational overhead and require extra resources for unnecessary complexity. Nevertheless, we see several trends that offset this rather gloomy picture and set the stage for creating a data management culture that can effectively respond to management information requirements. The first trend is a focused movement in our organizations toward greater efficiency and competitiveness. With this comes a willingness to consider a variety of strategic alternatives. These strategies include changes in support structures, processes, and responsibilities and often produce a rational analysis of data needs. Second, most business and institutions are undergoing migrations of at least some of their major data bases to new operational systems, often moving to increasingly distributed modes of operation. New development tools, as well as offtheshelf software, include structured methodology for clarifying an enterprise's data architecture, data definitions, and standard code usage. Since these methodologies provide the foundation for any executive information system, the migration projects present the opportunity to consider management's information requirements as part of the analysis. Third, there is recognition by small and large firms alike that there is a need for more participative management.[2] This compounds the numbers of managers who need to access and use data from across the organization. Thus, we see a momentum to also reach into other operational systems that are not immediate candidates for reengineering. All of this leads to a realization that there are ongoing data management activities that must be done to assure the availability of high quality information from all operational systems to support proactive decision making. There is much discussion in trade magazines and journals that couch these trends with such labels as reengineering, rightsizing, and total quality management. While each of these movements points toward improvement of the information support functions and spurs an interest in what can be done to quicken and maintain progress, the lack of a structured process for coordinating the various data management activities and relationships across the organization limits and jeopardizes the excitement and potential momentum. Still, there remains a multitude of potent opportunities within the existing culture to use traditional relationships and friendships to develop prototypes that push management's "hot buttons" and thereby create support for ongoing quality data management work. Since culture is dynamic and changing, it is imperative to proceed with a sense of urgency to produce quick results that demonstrate value to those who can benefit within the current culture. By producing relevant prototype examples that address real current needs, there is increased likelihood that quality data management issues and projects will be included as part of other evolving organizational changes, thus, formalizing the data management function. This paper presents paradigm for a new data management culture and then presents a leadership strategy for achieving that culture. A Data Management Paradigm There are three key roles in a data management paradigm that work together to create an organization's data management process. One role, located at the data source point, is the supplier. This role is generally played by a variety of functional offices such as finance or human resources, which have operational responsibility for segregated sets of organizational activities. Suppliers are the essential actors insuring the reliability of the data. They must control the random influences in the capture and storage of the data which can destroy the consistency, the stability, and the objective nature of attributes being captured. They are also responsible for documenting the capture and storage. Finally, the supplier must apply standardized recoding as part of creating cyclic extracts of the operational data for inclusion in the organizationwide repository, or data warehouse. These data custodial responsibilities are part of managing the functional area and are frequently delegated to designated system support stewards within the functional area. These distributed data administration activities are the basis of an organization'sdata management function. Next is the work of specifying standards that will allow integration from the various functional areas and will support data restructuring and delivery of entities as required by the numerous data users. This coordination is done by a central data management function, which plays the role of producer of management information. The producer is typically the custodian of a central data repository which contains important parts of the organizational data extracted cyclically and stored in a standardized form. In general, the central function is a bridging activity and this activity often includes running the organizational data warehouse with varying levels of responsibility for distribution. The person responsible for the producer functions may be referred to as the Data Administrator where there is a very limited distribution responsibility, or the Data Manager where the transformation of data into distributed information is also included. With the restructuring of data in a warehouse and the distribution of information comes responsibility for the internal validity of the data. This is the ability to interpret what the data mean and what the information created via analysis means. To the extent that delivery of information is included as part of the central function, the producers must also have the analytical ability to integrate qualitative and quantitative data and to determine likelihood of events. This may also involve manipulating the data to move it toward the format of the business rules for the customer to use in distributed decision making. Consequently, the producer is involved in quality assurance and implements checking mechanisms to determine consistency of business rules across supplier systems. The third role of user, or customer, involves using the information in some manner to identify the need to make a decision, select and evaluate alternatives, describe the situation, or defend and advocate previous decisions. The customer may want a subset of the organizational data selected based on predefined or newly identified key indicators. The level of flexibility that can be offered to the customer will depend on the sufficiency, relevancy, and timeliness of the organizational data warehouse. It is imperative that the customer be trained and learn how to evaluate the provided data in terms of sufficiency (some factors may need to be applied from external sources), relevancy (criteria for selecting subsets must be carefully determined), and timeliness (a determination must be made whether to use current data or trend data). These skills for interpreting and effectively using data are necessary for users to fulfill their data management responsibility, that of assuring the external validity. This represents the degree to which the data and information can be appropriately generalized to the specific situation. Since the specific situation, as well as the elements of the decision, are known only to the user, it is clear that the user or customer must carry this responsibility. Similarly, the user is responsible for monitoring and measuring the value of the information. This is often referred to as the construct validity of the data and consideration of the degree to which "asis" is becoming "tobe." In this regard, it is usually the user who is in the best position to identify changes in business rules that must be reflected by the organization's data. The user also needs to educate the supplier as to the true needs for data and the uses. This empowers the supplier to capture and store the most appropriate data. It is important to note that each of the three roles (supplier, producer, and customer) bring a different and necessary perspective to data management. The suppliers must be primarily concerned with a specific decentralized function. They have little reason to consider data requirements beyond those that directly support the function unless required to do so by cultural norms or rules. The customers' views may be somewhat generalized in terms of the subject matter of interest, but are most often as narrow as the most recent requirement for information. Customers function in distributed mode and generally interact with others only as necessary. They have limited views of the needs of other customers. The producer's view is of the organization as a whole. Producers cannot effectively assess either the "asis" or "tobe" infrastructures without significant interfaces with both the suppliers and the customers. Therefore, they must be central and accessible to both the suppliers and the customers. The producers are the central data management function. They coordinate and facilitate between suppliers and customers. Furthermore, they are the interpreters, providing translations to enable communication and transformation to relay understanding. Characteristics of Decision Making Information which reduces the uncertainty of the decision making or planning process is the goal of a successful data management function. In order to accomplish this, consideration has to be given as to what the information will be used for, the philosophy within which the information will be used, how the information is generated or created, who are the major players; and, finally, the current state of the campuses competing environment. Peter Ewell[3] identified five primary uses of information. Understanding the potential use of information is important as one creates information to help the decision maker reduce uncertainty of the decision. These five uses of informationrational decision making, problem identification, context setting, inducing action, and selling decisionseach require a different approach to the may in which the decision maker is supported. It follows then that the decision maker should be involved in the creation of this information. There are four primary decision making philosophies that we can anticipate being represented. The first is the political philosophy where the primary concern is with others' perceptions. In general, this philosophy is found most often with the CEO or president of your institution. His or her primary concerns are often centered around political forces external to the institution (legislatures, other external governing boards, or other community groups), or internal groups (deans, students, or faculty senate). This person usually has many demands on his/her time and will require a very discrete data elements focused on a specific topic. In other words, if meeting the need, a single number is preferable to a detailed accounting of all components making up that number. The implication is that the CEO has confidence in the decision support system and people to create reliable statistics. Again, as discussed below it will be important to involve him/her in initial discussions about the design of the data management environment. A second decision making philosophy is known as autocratic. In this case, the decision maker is primarily concerned with a specific and often personal agenda. Information requirements are characterized by a continuous flow of information that is focused on specific issues. In other words, the usefulness of the information you provide to this type of decision maker may be compromised if more than one topic is addressed in a single report. On a university campus, the person that most likely reflects this decision making philosophy is the dean. The dean typically does not have direct responsibility for an academic program or research effort. The primary role is to build a college along an agenda he or she has set. The managerial decision making philosophy is centered around the multitude of characteristics related to the specific program. The department head or chair typically reflects this type of decision making. In this case, the decision maker is interested in a continuous flow of information about any and all topics that are relevant to the programs in the department. The final decision making philosophy is known an collegial. In this philosophy, the context of primary concern is that of correctness of process. Typically, the faculty senate will reflect this type of decision making when working through an issue. The problem will be studied from every angle and analysis often conducted because of personal interest rather than from a notion that it would provide direct information to guide decision making. The type of information they require is often discrete but covering a broad range of topics. While in the above certain types of individuals on the campus have been identified as reflecting specific decision making philosophies, it should be remembered that at any given time a president could in fact use a collegial philosophy or any of the others if it suit a particular situation. Decision support must recognize the decision making philosophy that is in play in order to maximize the usefulness of their decision support information. Data Management Leadership Successful data management is dependent on advocacy and leadership spread across the roles of supplier, producer, and user/customer. It is predicated on nurturing new working relationships between people who previously did not communicate or were adversaries. Furthermore, there must be input representing all four of the decision making philosophies: political, autocratic, managerial, and collegial. What follows is a discussion of a strategy as a four step cycle (Plan, Do, Check, Act) that can begin to increase the robustness of data quality improvement in the current structure while establishing the foundation for the new formal and informal structures needed to deliver quality information. These steps amount to a strategy for "growing" a data management culture. Plan A critical step is to conduct a systematic evaluation of current data management and information support and develop a strategic plan for fortifying the process by which data are managed. Given the "new paradigm for management that encourages highly dispersed decision making,"[4] it would seem that one means for this activity to occur is to bring together the individuals involved in the development, analysis, and use of the data and to work them through a group process. Breaking the process into parts and involving individuals from various organizational units and perspectives is effective for quickly building consensus on immediate action steps. First, individuals in the group each identify issues. Small groups then cluster these issues, and barriers begin to surface. The next step is to focus on the root problems among the barriers, selecting one as a situation to address. This situation is then described as completely as possible and objectives and strategies to accomplish the objectives are formulated. Using such a process on a periodic basis accomplishes several things: 1) Awareness of problems and solutions increases and pervades the organization; 2) Doable tasks can be selected to improve information support and data quality; 3) Crossfunctional commitment and accountability are established; 4) Preceding successes are spotlighted in successive group processes. All of this means that incremental improvements are more likely to be sustained and resources allocated to make more progress. The plan must anticipate changing locations of decision making to refocus the type of information as well as the location of delivery. Do Actions taken to address the problems that become priorities as a result of the group process should be considered prototypes. This allows preestablished boundaries to be crossed as solutions are proposed. What works well can, and will, become accepted standard operating procedure. What doesn't work so well can be reworked with a greater chance of success given the new knowledge gained from the prototype. Although crossfunctional support and participation is key to the success of any innovation to improve information support, responsibility for coordination must still be clearly designated. Again, developing a prototype data management function is a good way to begin. For a select subset of the organization's data, develop a process to support: 1) reliable data capture and storage at the operational level; 2) internally valid data, extracted and integrated in a central data warehouse; and 3) externally valid data delivered to meet the needs of distributed management. Under the umbrella of centrally coordinated data management are the following functions and tasks which they must either do or coordinate. 1. Information Planning a. Work with operational area managers or custodians, system support stewards, and distributed users to maintain and share a model of data elements and their attributes important to the organization. b. Help anticipate and respond to users' changing information needs. c. Coordinate appropriate policies for population and use of the data warehouse. 2. Standards Administration a. Develop and coordinate implementation of standards for data warehouse data elements and codes. b. Work with distributed users, operational area managers, and information technology personnel in the establishment and implementation of appropriate policies for data management. c. Provide standards and appropriate documents for use by EDP audits. d. Maintain inventory of official definitions of codes and values for standardized data. 3. Operational System Management Support a. Assist managers of operational systems in extracting, standardizing, and providing data for the data warehouse. b. Assist managers of operational systems in developing and maintaining data elements and definitions. c. Provide limited training and advice to operational managers on developing and preparing individuals for datamanagement. d. Help operational managers create and coordinate user groups. 4. Team Building a. Coordinate identification of issues and strategies for dealing with data and information management. b. Support a quality development process with a baseline of standards and measures of improvement. c. Organize and provide administrative support to an information policy steering group which includes personnel from the information technology function, the central data management function, operational area managers, system support stewards, and distributed decision makers or users. d. Organize and provide training and administrative support to a group of distributed decision makers who use the data warehouse. e. Lead focused crossfunctional projects on data quality improvement. 5. Data Administration a. Manage the data warehouse, serving as custodian and system steward. b. Support creation of a consistent and usable set of data warehouse data elements. c. Where necessary, mediate user and operational area custodial concerns on codes and data definitions consistent with organizational authorities. d. Distribute information about the data warehouse and its use. e. Insure proper archiving and protection of historical data warehouse data. f. Act as the official source for the audit trail of changing codes and data element descriptions for elements in the data warehouse. g. Maintain currency of all policies on information management. 6. End User Support a. Provide examples of using the data warehouse to address concerns and problems across the organization. b. Help communicate user's information requirements to management and to the operational areas. c. Support user integration of local data with the organizational data in the data warehouse. d. Assist with migration of users' local data to the data warehouse when it is relevant to activities, analysis, or reporting that cross functional boundaries. 7. Technical Development a. Work with operational area managers or custodians, system support stewards, and end users to clarify technical needs. b. Support the information technology department in development, purchase, and use of prototype tools and products for managing and delivering data. c. Act as a clearinghouse for data management tools and related technology. Check For information support and data quality improvements to be sustained, it is critical that ongoing stewardship be executed by operational staff in the functional areas. It follows that there must be a mechanism for checking and verifying the adequacy of this stewardship. This amounts to assigned accountability for the following functions as a minimum: 1. Data standards and documentation. Work with the central data management function and others to develop and implement standards for selection, compatibility, integration and accessibility of organizationwide data. Incorporate and document implementation of these standards. 2. Data capture and storage. Assure reliable data capture, maintain list of allowable values, and archive data at standard cycles. 3. Data validation and correction. Implement and document validity checks and documentation in applications that capture, update, or report critical data. Develop and measure data quality and respond to problems by repairing the erroneous data, adjusting the processes that created erroneous data, and notifying impacted users of the corrections. 4. Data security. Implement and document access procedures that provide adequate protection as determined by management and monitor violations. Implement backup and recovery procedures that protect against threats to data integrity from system failure, faulty manipulation, or other disasters. 5. Data availability. Provide accessible, meaningful and timely machine readable data which clearly identifies capture and modification dates. Work with central data management to provide training and consulting on data use and solicit input for improving data quality and delivery. Act The work of the crossfunctional groups and the refinement of prototypes as a result of use and feedback continually nudge and encourage a data management culture. As pockets of support and commitment began to gel, the time is ripe to pour a firm foundation for data management. It is time to act....to formalize the coordinated data management function. Management of data and information support can become a critical part of the strategic core of the enterprise and those who are involved in these areas can becometeachers and leaders for others. The leaders will specify the people, activities, data, and tools that have potential to impact all facets of an organization and its outcomes. Some of the basics include: 1. Develop a data management policy. 2. Develop a criteria for selecting data required by distributed decision makers for inclusion in a data warehouse. 3. Follow the organization's lead, focusing on data from systems or processes that are targeted for reengineering. 4. Develop a standard for data definitions and descriptions for data warehouse data. 5. Select data management tools such as a data base system for the data warehouse, a complimentary data dictionary, guided user interfaces, report writers, etc. 6. Develop criteria, procedure, and process for refreshing and archiving data warehouse data and definitions. 7. Create crossfunctional teams from system migration workgroups to continually address data requirements and quality issues. These groups should include individuals who support and who use the operational transaction systems, individuals who work with data from these operational systems to provide it to decision makers, and individuals who use data provided for analysis and decision making. 8. Create a map of the information management structure with names and job descriptions on the various functions. 9. Develop training at all levels of the information management structure (operational area managers, system stewards, users) on the data management functions, procedures, and tools. 10. Initiate a program of professional development for those involved in data management and information support that sharpens project management, team leadership, and communication skills. 11. Create an environment and structure so the act of integrating improvement identifies the next set of activities where one can plan new ways to remove waste, scrap, and rework. Culture in Transition Progress in evolving a data management function or "growing" a data management culture will not always be readily apparent or easy to point out. Behaviors must change before the continuous improvement of data quality becomes a recognized organizational goal. The changes are subtle and incremental. A relevant model for the response of an individual to change is the KublerRoss sequence for grieving, i.e., denial, hostility, bargaining, depression, and acceptance.[5] This model is pertinent because it recognizes that change is first and foremost the loss of the "way we have always done it," and it is always traumatic because comfort zones are threatened. If we do not witness some of the behaviors in this sequence, it is likely that nothing is changing. There are several cultural indicators we should observe during transition that can alert us to the areas that require particular attention. These indicators can also help us assess whether we have truly evolved to a changed culture, or in the case in point, to a new data management culture. 1. Affective Orientation-Do people openly acknowledge and value working relationships with individuals across the organization? 2. Orientation-to Causality Do people claim ownership for problems as opposed to blaming others or the system? 3. Orientation to Hierarchy-Do people accept working with others at levels above and below them within the organization's hierarchy? 4. Orientation to Change-Are people willing to take risks and embark on new ventures? 5. Orientation to Collaboration-Are people receptive to others, seeking input and considering a variety of perspectives? 6. Orientation to Pluralism-Do people who represent different interest groups make an effort to make contact and communicate with each other?[6] To the extent that we can answer "yes" to each of the above questions, we are well positioned to evolve a data management culture that indeed embraces "new ideas for new challenges". We must remember that change is a process and subtle attitude changes signal progress. These include shifts from managing to leading, from control to coaching, from quantity to quality, from opinion to information, from resistant-to- change to open-to-change, from people-as-commodities to people-as-resources, from suspicion to trust, from compliance to commitment, from internal focus to customer focus, from individual to team, from detection to prevention.[7] All of these transitions are part of growing a data management culture. It is important to watch for these transitions and recognize them as gains in terms of awareness and commitment. They signal emerging pockets of support and fertile ground for adjusting priorities and strategy to use new allies as a reinvestment in the desired data management culture. Data management can no longer be relegated as a technical issue. There will need to be changes in our organizations. Matrix organizations and cross functional teams focused on specific tasks or problems must become common place. This places the data users and those who need information in direct contact with the operational sources of data. Those who produce information for decision making must shift from the roles of manipulating and analyzing data to roles of facilitator and trainer. Furthermore, those at the operational level will need training on the use of data and tools relevant to the new set of data stewardship activities which they will be expected to perform. The organization which tries to distribute data without data administration to define it and document it will fail. The organization which does not support the distributed data management function to push out the data and transform it into information will fall behind its more progressive competitors. The organization which does not restructure to be consistent with a revised decision structure will frustrate its employees and produce feudal turf battles. The organization which does not continually train its employees to manage and use information will make questionable decisions. We must step forward and nurture a data management culture. Deal and Kennedy suggest we must resist the temptation to roll up our sleeves and wade directly into the resolution of the problem as traditional managers if we truly want to encourage lasting change or transformation to a new culture. They suggest as an alternative the notion of a "symbolic manager" who recognizes "that the longerlasting solution is to rely on the culture to meander its way to a solution of the problem."[8] "New ideas for new challenges" amounts to creating a web of intelligence from the ideas and vision we have today. The web of intelligence is the new data management culture. It must be anchored by trained and empowered people who perform relevant and timely activities using sophisticated navigational and analytical tools on properly integrated quality data from both internal and external sources. "New ideas for new challenges" means change and the first change is at our own doorstep. 1 Radding, Alan, Quality is Job #1. In Datamation, October 1, 1992, p. 100. 2 Harper, Stephen C., Ph.D. The Challenges Facing CEOs: Past, Present, and Future, In The Executive, Vol. VI, No. 3, August 1992, pp. 10-11. 3 Ewell, P. T., 1989., Enhancing Information Use in Decision Making, NDIR, No. 64, Jossey-Bass 4 Quote by Don Tapscott, author of Paradigm Shift (1992, McGraw-Hill., New York). 5 Kubler-Ross, F., 1974. Questions and Answers on Death and Dying, Macmillan 6 Kuh, George D., and Elizabeth j. Whitt, 1988. The Invisible Tapestry: Culture in american Colleges and Universities. ASHE-ERIC Higher Education Report No.1. Washington, D.C.: Association for the Study of Higher Education. pp. 104-105. 7 Plice, Samuel J. Changing the Culture: Implementing TQM in an IT Organization. In CAUSE/EFFECT, Summer 1992, p. 23. 8 Deal, Torrence E., and Allen A. Kennedy. Corporate Cultures: The Rites and Rituals of Corporate Life, 1982. Addison-Wesley Publishing Co., Inc. p. 154.