Effective GIS Implementation

Effective Geographic Information System Implementation: An Investigation from Practice to Theory

Jeff Parkey

Clemson University

Abstract

Because of their wide application, geographic information systems have become a critical component of many organizations’ decision support system in recent years. Significant amounts of time, money and physical and human resources are often invested in developing and maintaining GIS programs. Despite the costs, agencies both public and private find great utility in the types of information and analyses that GIS bring. For these reasons, effective implementation of the technology within an organization is essential. Opinions of just how to achieve the successful implementation of a GIS are both vital and expensive. Obermeyer and Pinto note that “to better understand the management of geographic information systems within…organizations, a basic knowledge of organization theory and human behavior is essential.”[1] Of interest here then is an understanding of to what degree the prevailing GIS implementation practice corroborates concepts posited in organization and implementation theories. To this end, the methods of a leading implementation consultant were reviewed in light of applicable components of the organization theory literature. It was found that practice follows theory in a few important ways including: the notion of organizational validity, addressing human factors within the organization prior to the introduction of the technology, and also addressing the appropriate content and process factors within the implementation effort.

Introduction

According to Dr. John O’Looney, “…at least seventy to eighty percent of…local government’s work involves land or geographically related issues…”[2] Much the same could probably be said for governing bodies at other levels, as well as many private sector organizations. To this end, computerized geographic information systems (GIS) have become an increasingly integral part of organizations’ functioning. A GIS provides the ability to automate, develop, analyze, and output geographic and spatial inquiries of assets and interests, making it a powerful technology with broad applicability. Examples of its utility abound; organizations that deal with natural resource management, urban planning, facilities and infrastructure management, transportation, logistics and production, marketing and other business functions, delivery of government services, public health, national defense, and many other areas, have all adopted the technology. The microcosm, so to speak, of a typical local government setting can provide a further example of GIS applicability across an organization and in a variety of areas. Here, GIS can contribute to many of the bureau’s functions including engineering; environmental management; building regulation; land use planning and zoning; housing; fire, police, and EMS provision; public health; transportation planning; voter registration and elections services; property and land records management; school, legislative, taxing and other special purpose districting; economic development; public information dissemination; and finally just general mapping.

As a technological phenomenon, GIS extends into many of the concepts and constructs relevant to organization theory. It can be seen as an innovation, a tool for communication, a potential director of organizational design, an agent of change and development. It requires both a focused implementation and management effort.

It is upon this critical implementation stage that this discussion is centered. The theoretical literature includes numerous and significant contributions to the understanding of organizational management, however there are relatively fewer forays into that of implementation. This is not surprising, as a common understanding of implementing policies and programs is, generally, an elusive notion. Wildavsky and Pressman[3] provide what is still the most complete and accepted statement on implementation. Therein numerous contingencies and snafus, along with missteps in planning and a lack of foresight, thwart the efforts of a federal program established to assist with economic development among the urban disadvantaged. The overarching message of the work is the need for vigilance, preparedness, and careful planning when implementing policy. It will be seen that one who is implementing a GIS would do well in heeding this point.

In contrast to the literature that deals with implementation of other policies or programs, a growing body of literature concerning the implementation and related institutional issues of GIS seems to be growing. By some counts now an almost forty year old technology, it accords that a substantial literature has existed concerning the more technical aspects of GIS; the mechanics of its operations and the nature of its spatial computations. But with the ever increasing adoption of GIS technology by more and more organizations, recent interest in effective implementation, management, and other institutional, including even societal[4] issues, now surround it.

In this light, the works of Tomlinson, and Obermeyer and Pinto are particularly informative and pertinent to this study. Roger Tomlinson is a professional geographer who assists clients with the appropriate design and adoption of GIS. He is referred to as the “Father of GIS” due to his pioneering, large-scale work in the field beginning in the early 1960s. He has been at the forefront of professional, scholarly, service, and coordinating activities concerning GIS all over the world since that time. Nancy Obermeyer and Jeffrey Pinto are scholars in geography and management, respectively. They have written extensively on the subject of GIS, its management, and its consequence in organizations. Their 1994 book grounds the organizational employment and management of GIS within a context of social science, organization, and management theory, providing a reference for academics and practitioners both.

This paper will examine the prescribed implementation methodology that Tomlinson provides to his clients in juxtaposition with the body of theoretical concepts presented by Obermeyer and Pinto in an attempt to discern if today’s state-of-the-art implementation practice is consistent with the observations made in the literature of organization theory. The search for such a consistency helps to legitimate and/or reconcile the approach of the professional guru and the evidence found by the theorist. It can help to validate the aim of theory, generally.

Current GIS Implementation Practice

Dr. Roger Tomlinson is a world renowned expert in geographic information technology. President of Tomlinson Associates, consulting geographers, Tomlinson has guided many public and private organizations the world over through the implementation of their GIS. Tomlinson’s clients have included numerous agencies within the U.S., Australian and Canadian federal governments, The United Nations, the World Bank, and many local government agencies and private firms. At the heart of Tomlinson’s implementation method is a ten-step process that he calls the “planning methodology.” Within its steps the planning methodology covers preparatory tasks, system requirements, intra-organizational issues, costs, and other important concerns. Tomlinson’s biggest claim however is the pre-planning effort that must take into account personnel issues, internal politics, and the like. The steps in Tomlinson’s planning methodology are listed here then briefly discussed below.

Steps in the Tomlinson Method for GIS Implementation[5]

Consider the strategic purpose
Plan for the planning
Conduct a technology seminar
Describe the information products
Define the system scope
Create a data design
Choose a logical data model
Determine system requirements
Perform benefit-cost, migration, and risk analyses
Make an implementation plan

In the first step of the planning methodology, consideration of the goals, objectives, mandates, and overall purpose of the organization is advised. In so doing, the planning and implementation process and its result should square with the organization’s mission.

What is meant by “plan for the planning” is to understand, and to make the organization understand, the requirement for resources, namely time and people, to simply initiate the implementation of the GIS. Basically, the organization must sanction the time and effort being spent by employees in planning for a GIS; if it does, commitment to the GIS on the part of the organization is understood.

Holding a technology seminar is an integral part of the implementation process as it defines exactly what a GIS is to the organization it will serve, and also defines to the future GIS what the demands placed upon it by the organization will be. The users or “clients” within the organization are given the ability to voice what it is that they wish to gain from the GIS specific to their job or role in the organization.

Exactly what the products will be that address the needs of the users found during the technology seminar are outlined in step four. Follow-up, focused, individual discussions with each participant from the technology seminar yield information about exactly what they do in their job and therefore what is needed, in terms of products or deliverables, from the system to assist them.

In step five, the scope of the GIS is determined. Essentially, the data that will be needed to produce the information products is assessed with this step. Exactly what data is needed, when it is needed, and how much of it, are addressed.

The database design pertains to many of the important technical details about the spatial data that will be used in the GIS. Cartographic details, data capture and conversion, standards for the data, and other similar technicalities are dealt with in this step. Several independent and also interrelated data layers are defined in the database design.

The choice of a logical data model for the GIS reflects the nature of the geographic features that are dealt with by the organization. There are a few kinds of automated data models for handling spatial data, and based on the complexity or simplicity of the organization’s analyses of its geography, tolerance for error, accuracy and update needs, and other issues, one or many models will be used in the system.

The system requirements are the hardware, software, user interface, networking, storage, and other physical resources that will be needed to handle the information products, system scope and design, and data models defined by the above steps.

Until step nine, everything has dealt with the “what”; that is, “what” exactly is the organization going to implement. Examining costs and benefits and the migration and risks involved with GIS implementation are the “how” of the implementation process. Legacy systems, staffing and training, funding, and legal issues are outlined here.

Finally, a report is produced that should communicate all of the above findings to management. The organizations objectives, the information requirements of the system, its design, implementation recommendations, timing, and funding should all be unified within the GIS and expressed in this implementation plan.

Tomlinson points out that throughout the course of working through the steps in the planning methodology, continual communication should occur between those implementing the GIS and those who are the supervisors of the implementers. In so doing, continual support for the implementation effort should be maintained, and there should also be no surprises by either side as events unfold.

Tomlinson’s planning methodology, particularly through a few key steps, becomes a very strategic, well-rounded course of action that demonstrates forward thinking and an interest in consensus-building. The middle stages of the method involve technical details specifically concerned with the hardware, software, and data inputs; the physical components of the technology. No GIS implementation plan can exist without these considerations. What is different here though is that after considering all of these technological factors, many GIS implementations plans would stop. Tomlinson’s planning methodology however takes into account a wholly other set of issues; those of the context of the organization. Further, the methodology does not just present these issues; it also provides recommendations to the implementer for how to address them. By insisting that the organization’s purpose and goals be understood, the likelihood of the alignment between the GIS and the needs of the organization is better guaranteed. By getting the “OK” from upper management to plan for the planning of the GIS, an important measure of senior support is gained. This basically tells the implementers that they can proceed with making plans for the technology, and that the technology is wanted. Through the technology seminar and also the definition of the information products, the buy-in of the organization members, those who will actually use the technology and its outputs, is obtained. Their understanding of the new system is ensured and their voice is heard concerning the system’s design and the expectations for it. Finally, by broaching the subjects of funding, procurement, staffing, changeover from the present system, legalities, and other related issues, the final steps ground the implementation effort in everyday, functional reality.

The Tomlinson method seems to be one that is complete…even rational, to some organization theorists, depending upon what might be included in defining that term.

Implementation Theory

Obermeyer and Pinto are quick to point out that the acceptance and utilization of technological innovation within an organization is most critically linked to institutional and behavioral issues. Innovation in an organization is often experienced as organizational change. Its introduction and adoption can therefore be controversial. Operatives within the organization that interface with technology to various degrees and in various ways, and also whose standing within the organization varies, may come to the technology with different expectations, needs, and even trepidations.

Implementation is defined as “a new…system, program, or model that has been accepted by organizational personnel, the results of which…change the decision-making process…”[6] This notion of acceptance is crucial to the successful implementation of innovation. Apparently operatives must understand the technology, its use and/or usefulness, and its benefits. Indeed, Schultz and Slevin[7] point out that the support of the organization and its endorsement of the technology is won not through resolution of the mechanical intricacies of operating the technology, but instead by appealing to the concerns of the organization’s membership.

The technology’s organizational validity[8] is related to the acceptance idea. An innovation is valid for an organization if it meets with the needs of users, clients, and/or customers. Further, it helps the organization to fulfill its purpose or promise, its mission and operations. Acquiring technological innovation for its own sake would be a flawed policy, according to the theory. To the extent that application of the innovation squares with the organization’s activities, it is expected that the innovation will be accepted and will be viewed as satisfactory.

Theories of implementation also include the concepts of implementation content and implementation process. Content and process concepts basically address the “what” and the “how” of the implementation. Implementation content models include ideas and techniques that are concerned about the actual innovation itself, such as goals, resources, support, scheduling, competency, communication, and responsiveness. Process models of implementation capture the way in which the implementation proceeds. These are the tactics that will be used to do the implementing, the actual actions that are taken. In viewing the implementation effort for its content and process, Obermeyer and Pinto find that

While each method is useful, neither offers a complete picture. A thorough approach should identify both the key decision factors in adopting geographic information technologies and the processes by which the implementation occurs.[9]

Both content and process factors of implementation should be examined if a complete implementation is desired.

In terms of implementing a GIS, or any new policy or program for that matter, organization theory that involves innovation, change, and development is useful in guiding action. Obermeyer and Pinto note that

The concept of implementation in the context of organizations may be viewed as a change phenomenon or a process for creating organizational change. Initially the problem of implementation was discussed in the context of frequently ineffectual attempts on the part of operations researchers and management scientists to generate enthusiasm for and use of a myriad of new organizational innovations intended for use by practicing managers. More recently the problem of implementation has been represented as the frequent failure to create some degree of desired organizational change through the introduction of a new information system, program, or model.[10]

Their comments point up the difficulties inherent in introducing an innovative concept or system into an organization, thereby changing the organization in some way. The organizational change and development theorists basically describe change in an organization as being “planned” or “unplanned”.[11] This concern for some amount of planning as included in or prior to the implementation of a GIS seems to be a constant thread throughout much of the findings of theorists of implementation. It seems obvious that with greater degrees of advance preparation and consideration for how things will go, a more successful and effective introduction of GIS into an organization can be predicted, versus simply jumping right in and instituting GIS in a more wanton manner.

Discussion

Tomlinson’s planning method is an interesting encapsulation of many of the aspects found in the organization theory dealing with implementation, innovation, and organizational change and development. By understanding from the outset what the goal and purpose of the organization is, the role that the GIS will play in that organization is better understood and the arrangement of the GIS to best function in this role for the organization is made more possible. By ensuring that resources can be dedicated to a planning effort for the GIS, implementers of the technology get the blessing of those who will be making the ultimate decision to go with the technology or not. By engaging those who will be the users of the system to determine their wants and needs, ground-level support for the innovation along with critical advice on how to structure the system is gained. This advice coupled with an extensive survey of the required physical and data infrastructure, tells the implementers just what exactly they will be implementing. The last two steps in the planning methodology tell the implementers how they will carry out the implementation, as they relate to perfunctory aspects of system introduction like procurement, funding, migration to the new system, and personnel. By investigating all these issues prior to purchasing and setting up an organization’s GIS technology, the planning needed to better ensure success is present. Tomlinson himself claims

…the evidence shows that good GIS planning leads to GIS success, and the absence of planning leads to failure…you must integrate sufficient planning into the development of your GIS; if you don’t, chances are you’ll end up with a system that doesn’t meet your expectations.[12]

All of the ideas above are found to be important by the organization theory.

In Tomlinson’s words, “GIS is a particularly horizontal technology in the sense that it has wide-ranging applications across the industrial and intellectual landscape.”[13] In much the same way that so-called “flattened hierarchies” and even traditional information technologies work, so too does GIS work. The GIS creates communication and interaction across the boxes on the organizational chart through data sharing and cooperative action on GIS projects. Beyond this though, a GIS could be seen as a “leveler” within an organization; or better a “unifier”.

The introduction of a GIS within an organization exposes flaws and gaps, in the communication strategy, in the hierarchy, in what are or ought to be common resources. At the same time however it reveals common needs and desires, and common useable sets of data or pieces of information. Because of its broad application, that is its ability to work for and into so many issue areas, it will naturally level, unify, and connect, many of the operational areas within the organization. If one wished to know what each department or functionary of an organization actually does, and what they need to do it with, the first place one might look would be to the head of the organization. A much better place to look might be the head of the GIS. Unlike more traditional information technologies that are static, GIS can be fitted to the specific needs of the operatives by acquiring and providing the particular geographic data that they use in their work, and in enabling analyses and outputs specific to the operative’s interests. The leader of the GIS has to by necessity know these details about the organization. The interrelatedness of the departments or divisions of the organization becomes amplified with the insertion of a GIS. It may even be felt that the organization becomes better connected to itself or better connected within itself, or more coherent in performing whatever its task is. An unforeseen bi-product, however, can be the exposing of internal political issues engendered by this new “closeness”.

In this author’s experience, both the practical and theoretical ideas that have been presented here have indeed been encountered in the workplace. In two different bureaus’ GIS departments a documented “GIS plan” did exist. The document included information regarding design and implementation that corresponded to a considerable degree with the steps and reporting that Tomlinson advises. One of this author’s tasks was to keep the GIS plan document current as new events or information arose.

Also in this author’s experience is the undeniable fact that resistance to innovation is most strongly rooted in organizational and not directly technical aspects. It was the somewhat unfortunate experience that in some areas of GIS implementation, insufficient information product definition, as Tomlinson would put it, had occurred. Ever after, the components of the system that were allegedly designed to meet a particular department’s needs were not only misunderstood and underutilized, but also vehemently opposed or ignored. On numerous occasions these personnel complained that they were never “in the loop” from the beginning and this particular component of the GIS did not meet the specifications that they would have liked to had the opportunity to put forward. Yet other operatives were simply unconvinced of the need to accept a new technology and change the way that they had always done their work. Any benefits to the organization of this innovation were lost on these workers, who notably were much closer to retirement than further from it. In the cases of each “resistance”, it seemed the problem was less one of intimidation by the technical aspects of a new computer system; where operatives were either required or eventually voluntarily participated, in time they developed the technical skills necessary. Interestingly, this technical development did not seem to change much in their overall attitude toward the “interloper” that was the GIS.

It is of course with something of a grain of salt that this investigation can use the actual practices of a paid consultant to test theory. In the instance of empirical research, a more rigorous effort to eliminate bias ought to be preformed – perhaps by obtaining the opinion as to the success or satisfaction with their GIS implementation effort of an agency that had been a client or had used Tomlinson’s method. For a consultant who is being paid to implement a GIS, it seems unlikely that he would halt the process if something within the planning methodology were not up to snuff. Nothing could be admitted to be an impediment of the implementation. By virtue of this, there would never be anything but successful implementations – in the eyes of the consultant, of course. The client agency might think otherwise. To meet the critical standards of research, hypothesis testing, theory-building, etc. it would be the comments of the client organizations that would probably be most efficacious.

So is the organization theory concerning implementation, innovation, and change corroborated by the practice of today’s guru in the GIS field? The answer is yes. It may, though, be a qualified yes. There is something of a chicken and egg problem that must be admitted in attempting to validate theory with practice, and vice versa. Simply put, all theories are built upon what is observed from the field. The commonalities among these observations are knitted together to achieve a formula or guide for predicting what will happen when similar practical applications surface. As such, the formula that is the theory will never predict everything that can or does happen just right. Nor will it be applicable in every single case. But some of the time it will of course be useful because of the manner in which it was created; it is rooted to some degree in the practice itself. The same of course stands for practice following theory. Real-world applications in the field will not always follow just what the theory suggests, but sometimes and to some degree it will; again, it must, because the theory was built form watching what goes on in the real world. In the face of all this then the real question may be: is it a valid question to ask if practice follows theory?! Here the answer is a definite yes. For organization theory or any other discipline it is always a useful exercise to reinforce the connection between theory and practice, and to validate the purpose and utility of theoretical enquiry generally, to students, scholars, and practitioners.

Implications for Policy and Future Research

To frame this issue in a public policy context then, it may be interesting to ask: for what reasons might it be important, outside the organization itself, that a successful or proper GIS implementation occur? In the case of a public organization, probably the most clear policy or political implication is due to the system’s funding. That is, it is most often the public taxpayers who fund a GIS, be it for a city, county, state or other administrative or regulatory agency. Therefore, the taxpayer has a vested interest that the system works appropriately. Indeed, the taxpayers probably have more of a vested interest than they know when it comes to the data, uses, and capabilities of a GIS.

It has already been noted that more and more organizations in all sectors, all over the country, are adopting GIS to help them in their work. This accumulation of GIS users has put an ever greater need on standardized data and processes as more spatial data is used and shared and more GIS output is depended upon as a guide to decision-making. There are numerous examples of standardizing and coordinating efforts underway presently. A significant one is the Federal Geographic Data Committee (FGDC)

…a 19 member interagency committee composed of representatives from the Executive Office of the President, Cabinet-level and independent agencies. The FGDC is developing the National Spatial Data Infrastructure (NSDI) in cooperation with organizations from State, local and tribal governments, the academic community, and the private sector. The NSDI encompasses policies, standards, and procedures for organizations to cooperatively produce and share geographic data.[14]

Also of note is many states’, including South Carolina’s, efforts to consolidate and coordinate geographic data in a single location for end users to access via the internet. It is obvious that such a program would take a high level of coordination, management, and implementation success as well as ensuring standardized data with consistency and integrity. The benefits are great however, not the least of which are one-stop data shopping for the end user, and the use of consistent, congruent data across projects.

Finally, the effective design, implementation, and management of a GIS is important to policy as it enables the soundness of information that is used in decision-making. Of all the variables that go into the decision-making process – the public good, politics, power, etc. – the information that is used is of equal if not greater importance than any of these others. Interestingly, the way that a given body of information is or exists in context can be an important enabler, but also delimiter, to the decisions that are made. Recent, compelling examples of this situation have been see of the world stage as the United States decided to go to war in Iraq based in part on intelligence that it appears may have been incomplete or at least malleable. Particularly when information is used to make decisions concerning the public interest, it should be a goal that that information be as good as it can be. This would include the implementation of the systems that handle that information.

It would be remiss to not suggest a few related directions for further enquiry. Some might include: assessments from testing the level of overall satisfaction that an organization feels toward its GIS; the data and information acquired from GIS and remote sensing as objects in the public domain…how are they used, access to them, the sale of them, the security of this data and, moreover, its use for security or safety purposes; when is a GIS needed by an organization, really, and what are the reasons and justifications given?

The onset of the use of GIS and other advanced information technologies within organizations, especially public ones, brings with it a host of issues and opportunities for those organizations in pursuing their goals. Organization theory can provide lessons and advice that can assist these organizations in the practice of effectively incorporating these innovations into their operations, purpose, and mission.

[1] Obermeyer, Nancy J. and Jeffrey Pinto. 1994. Managing Geographic Information Systems. Guilford Press, New York, New York.

[2] O’Looney, John. 2000. Beyond Maps: GIS and Decision Making in Local Government. ESRI Press, Redlands, California.

[3] Pressman, Jeffrey L., and Aaron Wildavsky. 1984. Implementation, 3^rd Edition. UC Press, Berkeley, Los Angeles.

[5] Tomlinson, Roger. 2003. Thinking About GIS: Geographic Information Systems Planning for Managers. ESRI Press, Redlands, CA.

[7] Schultz, R.L., and D.P. Slevin. 1975. Implementation and management innovation. In Schultz and Slevin, eds. The Implementation of Management Science. Elsevier. New York.

[11] Gortner, Harold F., Julianne Mahler, and Jeanne Bell Nicholson. 1997. Organization change and development. In Gortner, et al, eds., Organization Theory: A Public Perspective, 2^nd Edition. Wadsworth Group, Belmont, CA.