The Interaction between IS and Business Units
in Setting IS Standards

by Judith Gordon and Steven Gordon

CIMS Working Paper Series 97-07

Judith R. Gordon
Associate Professor of Management
Boston College
Chestnut Hill, Massachusetts

Steven Gordon
Associate Professor of Information Systems
Babson College
Babson Park, Massachusetts

1997
Center for Information Management Studies (CIMS)
Babson College
Babson Park, Massachusetts 02157-0310
Phone 781-239-4531 FAX 781-239-6416


Abstract

 

In this paper we examine the process of setting and enforcing standards for information technology and systems in companies having multiple business units. Unique challenges to this process include managing the extent to which standards should be set; dealing with language, cultural, and geographical differences; the choice of centralization or decentralization of standardization; the alignment of IT and organizational strategies and their impact on standards; and the role of organizational structure in handling standards.

Within these challenges of the marketplace, companies with multiple business units have demonstrated a focus on the primacy of the business units in the delivery of information systems services. This paper attempts to delineate the role that business units play in the standard setting and enforcement process. We report the results of a pilot study that, among other issues of IS service delivery, examined the nature of the setting and enforcing of standards within eight Fortune 500 manufacturing companies. The study answers three questions: (1) What prototypes describe the IT-business interaction with regard to standards? (2) What are the advantages and disadvantages of each prototype? and (3) What are the outcomes of these approaches to standard setting? This paper concludes with a set of implications and conclusions that highlight lessons for CIOs and general managers regarding developing effective relationships in the setting and maintaining of standards.


The Interaction between IS and Business Units

In Setting IS Standards

 

Introduction

Companies that consist of multiple business units face an array of challenges. Perhaps none is greater than creating compatible information systems among units that differ significantly in the products and services they offer, the processes they use to create value, and their culture, experience, and location. One way to achieve compatibility is to set standards that all business unit can endure and even support. In this paper we examine the process of setting and enforcing standards, which we define as "sets of rules or policies governing the characteristics of data, software, and/or hardware that an organization may purchase, develop, or maintain."1

We begin by describing the challenges posed by the multi-business-unit context. Then we describe the importance of business units in IS service delivery. We continue by describing our research methodology and the companies we studied. Next we provide results that answer the following questions: (1) What prototypes describe the IT-business interaction with regard to standards? (2) What are the advantages and disadvantages of each prototype? and (3) What are the outcomes of these approaches to standard setting? We conclude the paper with a discussion of the implications of these prototypes for introducing an effective standardization process in organizations.

The Challenges Facing Companies in Handling Standards

The setting and maintaining of standards occurs within the organizational and competitive context that multi-business-unit companies face. This context provides five major challenges for delivering IS services in general and dealing with standards in particular.

  1. The extent to which standards should be set. Although most practitioners and researchers agree that some standards should be set, the extent depends on the assessment of tradeoffs between benefits and drawbacks.2 For example, while standards minimize duplication of software development, they also reduce flexibility in applications.3
  2. The degree to which IT service delivery should be centralized or decentralized. Research has contrasted the control, efficiency, and economies of scale created by centralization with the flexibility, empowerment, service-orientation, and effectiveness in meeting individual needs of decentralization.4 The setting of standards operates within this general paradigm for IT service delivery. Organizations must choose the extent to which centralization places control of standard setting within the hands of a few top general managers or top-level IS personnel as opposed to allowing individual business units and even groups within those units to determine standards that meet their unique needs.5
  3. The role of organizational structure in prescribing the handling of standards. The extent to which organizations introduce mechanistic structures, with extensive centralization of decision making, unitary chain of command, extensive division of labor, and high specialization, as opposed to more organic structures, with decentralization of decision making, multiple or more flexible chains of command, division of responsibility based on expertise rather than job title, and less specialization, will have implications for the setting of standards. As organizations increasingly adopt new organizational forms, the approach to standard setting may either support or hinder autonomy of decision making and rapid responses to competition.6
  4. The availability of standards that span cultural, language, and geographical boundaries. The setting of standards is particularly problematic in companies whose business units span international boundaries and for those whose business units are geographically based. Operations in countries such as Japan, where national culture, standard ways of doing business, and a general isolation and autonomy differ from those in the United States and Europe, hinder the establishment of standards.7
  5. The impact of global strategy on standards for companies that operate internationally. Companies use one of four strategies - multinational, global, international, and transnational -- when operating in the global arena.8 Research has shown that companies that differ in their internationalization strategy differ as well in the structure of their information systems services delivery.9 Similarly, they vary in their need to share data and applications, the locus of decision making, extent of control by the parent over the subsidiary, and nature of global integration. In line with these structural variations, they likely also differ in their approach to standard setting and enforcement.

The Primacy of the Business Unit

The context provided by these challenges interacts with a strong focus on the centrality of the business unit and its importance in the delivery of IS services.10 This emphasis on the business unit stems from the desire of companies to adapt and respond to customers' requirements, perform effectively in a rapidly changing and challenging environment, and sustain their position in light of other competitive pressures. At the same time, focusing on the needs of business units runs the risk of removing IT expertise from decision making unless mechanisms exist to ensure its input. It is essential, therefore, that IT managers and staff understand their organization's business so they can successfully relate the implementation of IT activities to the business's objectives. This study examines the implications of the centrality of the business unit for the design, delivery, and maintenance of IT standards.

The Study Methodology

This study uses a qualitative research methodology in which detailed interview data were collected and analyzed to answer the three research questions. We interviewed top-level information officers (e.g. vice president of information technology, corporate CIO, senior manager of information management) of seven Boston-area manufacturing firms and one Boston-based subsidiary of a U.S. multinational manufacturing firm. Firms were selected based on the researchers' ability to get access to senior IT managers, rather than being randomly chosen. Because this was a pilot study, we selected participants who we felt would be candid in discussing the issues around standardization in their companies. We asked each person a series of questions, similar to those shown in Table 1, designed to help us understand the overall structure of their company as well as the structure of the delivery of IT services. Each interview lasted between one and two hours and the responses provide the basis of the analysis in this pilot study. We had each interview transcribed (with the exception of one that had poor audio quality), and also took detailed notes during the interview. We read each interview and identified any responses related to standardization; we then identified the topic of these comments and recorded the data by topic for each interview. In this way the data suggest the categories and their dimensions; the data were then organized according to recurring themes.11

Table 1
Interview Questions

1. What are the functions of the corporate IT group?

2. Where is the primary location of IT resources (geographical, business unit, group level)?

3. What type of lateral linkages exist between IT and business employees?

4. Who has responsibility for standard setting (IT or business unit)?

5. What is the nature of decision-making within and for the IT function?

The eight firms differed significantly in their industry, size (in sales dollars), overall structural configuration, and the primary location of IT resources, as shown in Table 2.

Table 2
Description of the Organizations

 

Company A

Company B

Company C

Company D

Company E

Company F

Company G

Company H

Industry

heavy manufac-turing

consumer goods

consumer goods

consumer goods

health care related

computer-related

computer-related

computer-related

Sales Dollars (in billions)

0.6

2

3

6

2

32

14

2

Structural Configur-ation

matrix of world-wide divisions and functions

functional

product-based

product-based divisions

multi-divisional and multi-product

product-based global groups

product-based divisions

functional

Location of IT Resources

three geograph-ical support locations

senior IM functional area support and 30 geograph-ical IT groups

corporate and geograph- ical groups

corporate IT, business unit IT groups, and geograph-ical groups

corporate and divisional It groups

product IT groups

business unit IT groups

corporate IT

 

Even within industries and for companies of comparable size, differences existed in the delivery of IT services. For example, the three consumer-goods companies (B, C, and D) had two different types of structures (function and product); Company B located its IT resources within the functional areas and geographical areas; Company C located its resources in corporate and geographical groups; and Company D located its IT resources in corporate, geographical, and business groups. Companies B and H, which were similar in size, both had functional structures, but Company B located its IT resources in the functional and geographical groups, whereas company H located them in corporate headquarters.

Dimensions of Standardization

The data suggest that the organizations differ in their approach to standardization along five dimensions: (1) drivers of standard setting, (2) extent of centralization and decentralization, (3) influence of IT and business units, (4) decision-making approach, and (5) type of oversight. In the next sections we describe these dimensions and the nature of the organizations' standard setting along them.

Drivers of Standard Setting

Two major factors affect the process of setting standards: technological knowledge and business needs. These drivers influence the extent of centralization and decentralization, the role of IT and the business units in setting standards, the decision-making approach, and the nature of oversight for the process. Historically, technological knowledge drove the standard setting process because people responsible for setting standards did not believe that business people had sufficient technical expertise to participate in the process. This belief still exists in some companies such as Company A in our sample.

We selected Oracle, we selected Unix, we selected relational. Very few of our business people in fact could contribute effectively to technology standard setting.

More recently, the growing emphasis on the primacy of business units has caused the business needs to drive standard-related activities. Where top management believes that IT can truly understand the business needs of the company and parlay this knowledge with their technological expertise, handling standards tends to remain in IT's domain. When top management believes that only the business units can truly understand their business needs, responsibility for standards shifts, either partially or totally to the business units. In Companies B and D, for example, the predominance of local benefits likely explains the reluctance of business units to relinquish responsibility for standard setting and enforcement to IT.

What I'd say historically has happened is that these teams have enough autonomy so that if they select something that doesn't fit well in the architecture, unless it is a grievous error, they have been strong enough to drive that project through. So we have ended up with a one of these and one of those kinds of environments.

In Companies F and G, in contrast, business needs worldwide drive the standardization process, and result in joint standard setting by IT and the business units; IT brings a global perspective and is viewed as understanding the complexities of technology integration in a global company; the business units ensure that the standards meet their unique needs.

Extent of Centralization and Decentralization

Centralization versus decentralization refers to the amount of discretion individual business units and subunits have in selecting or changing standards. Centralization leaves standard-setting in the hands of the top corporate executives or the top IS executives, resulting in greater consistency and potential economies of scale. For example, Company C retains control of standards centrally through IT enforcement. As the CIO described a situation where he successfully refused to yield to business's desire for a special software platform:

He [the head of a division] called me up one day and said, "I'm ordering you to buy ACS; I'm tired of this, I'm not getting the support I need. There is a good solution out there that we're using in San Diego called ACS, runs on an AS 400. I'm ordering you to buy this...everybody's going to be using this." I said, "I'm sorry, you can't order me to do it." "What do you mean I can't order you to do it?" And I said, "I don't take orders from you. And it's the wrong decision for the business...I am willing to increase the support to [your business]; I'll do whatever it takes; I'll be accountable for solving the [business's] problems...you're not going to tell me how to solve the problem..." What was really important is supporting the business. But it really was about continuing to hold on to control. Don't tell us how, tell us what your issues are and we will figure out what is the best way to do it.

Decentralization of standard setting more explicitly acknowledges the centrality of the business units and their needs, allowing standards to support the customer orientation and flexibility expressed by individual units. In Company B, for example, individual business units can deviate from the set standards as they desire. Although there is an attempt to bring more centralization to the standards processes, the culture of this organization discourages such movement toward greater standardization.

In the last three years, I think we have made some major strides toward bringing the total business organization into an awareness that there is a value in doing things in a much more integrated fashion. Again, I think we have had a culture that has been very local-centric and I think that is changing a lot. You like to think, we [IT] finally won that and got them to understand, but it isn't that at all -- the world has changed.

The extent of centralization versus decentralization is increasingly in flux, as market, shareholder, and other environmental demands prompt corporate restructuring. Company D, for example, describes itself as recentralizing, moving from decentralized standards to more centralized ones as part of their attempt to go global in their operations. Currently they use common software, where possible, but still lag in having common processes:

We have a common systems group and a corporate IT group, but what we find is that their focus is supporting a common application and not really supporting common processes...And we'll end up providing technical support for the package, but that doesn't mean that we have agreements between Singapore and Australia and Japan that they're going to put a common...process there. So our common systems group today, what they have to say is, okay, at least we're using common software and we're trying to implement that in a consistent fashion.

 

Influence of IT and Business Units

IT and business units generally play complementary roles in the standard setting and enforcement process. Our data suggested four types of roles. First, IT can unilaterally elect and enforce standards and the business units then live with them. At Company A, for example, where technical knowledge of IT drives the standardization process, IT has sole responsibility for standards.

It's fair to say that we absolutely select and enforce the standards with minimal business input... Technology standard setting is done essentially by MIS.

Second, IT can select and enforce standards, but give the business units some discretion in their timing or offer the possibility of overriding selected standards. The CIO of company C uses a traffic light analogy in describing the discretion given to business units around standards:

If somebody came to me and said I'd like to buy a personal information manager. I can get all hung up about why don't you use Lotus organizer, why can't you standardize on ACT, why can't you...Yet that's not where there's any return to the business. I mean people can make these kinds of decisions on their own. Now would there be a possibility of saving some money for the corporation saying could we organize a volume discount? ... [Perhaps,] but frankly the cost of trying to administer that is not worth the return to the business. You just simply say, it's a green light; I don't care what you do. If we said, gee, we know that within SAP we're going to be able to handle material purchasing or material bidding within the Far East, but I can't get there for a year and half. If you can find something that satisfies you on the short term, you know it's an area that I would consider yellow....There are other cases where you say it's an absolute never-ever can you [do it] -- it's a red light.

Third, IT and business units can jointly set and enforce standards. As IT joins with the business units in setting standards, the process reflects a greater concern with business needs and a perception that IT lacks sufficient information about business needs to meet them effectively. In Company G, because the standard setting process is quite extensive, and gives individual units opportunities to veto proposed standards, once agreement is reached the business unit needs to live within the standards:

So essentially the Chief Technologist in each business unit has a subcommittee that discusses what standards we ought to be using in a company. Obviously once initialized it doesn't change dramatically every week. They recommend to the whole staff what standards we ought to be using as a company. If any individual in my [corporate IT] staff doesn't like the standard, they can veto it, in which case it goes back to the organizational entity for review. So at that point you've got a buy-in process that is pretty hard to have anybody come forth and say I can't conform to that standard. And if they believe that the market or the technology has moved on and therefore we ought to be looking at another standard, then all they have to do is have their person go into the group and start running that through the process.

Fourth, IT can recommend standards and the business units can then select and enforce the standards. When IT recommends standards to the business units, the business unit makes the ultimate decision around standards. At the extreme, as for Company B in the past, when IT merely provides support for standards selected by business units, great discretion in the choice of standards results in tremendous variety throughout the organization.

Decision-Making Approach

Decision making around standards ranges from directive to collaborative approaches. Either IT or business units can take a directive approach. In the first case, this generally stems from the agreement that IT's technological knowledge allows it to preempt business units in standard setting.

Alternatively, IT and business units work together in some negotiated, collaborative, or consensus-reaching style in which they attempt to share their needs and attain buy-in. In Company C, this buy-in revolves around negotiating the timing of meeting new standards. Company G takes a collaborative approach to standard setting, with buy-in the ultimate outcome. Company D attempts to attain consensus about standards within the context of meeting individual business unit needs while at the same time developing an integrated, global structure. In company E responding to business needs occurs by bridging diverse standards in different parts of the organization:

Rather than try to get everybody to change to the same network, we have done some things to allow interoperability between the two or three mail systems we have. We have been able to allow CCMail to talk to BanyanMail just as if they were the same. And now we're implementing a Microsoft exchange pilot whereby CCMail, MSMail, and BanyanMail, all operating as SMTP-type mail systems, including some VMS mail, will all be able to inter-operate. So our only rule is...you have to be the same mail system in a division...Someday we may have a common mail system....There is no rush to get there because it doesn't cost us very much more money to have this bridge in the middle. So that is sort of an example of how we mitigate some of these issues.

Oversight

Oversight groups can allocate responsibility, accountability, and resources for standard setting and enforcement. The nature of the oversight reflects the relative importance of IT versus business unit in the standardization process. Companies had one of three types of oversight groups: (1) an IT steering group, (2) a business steering group, composed of top corporate executives or line managers within businesses, or (3) mixed groups, composed of some executives, some divisional leadership, and some IT representatives.

Companies that clearly valued IT's expertise used planning groups within IT. For example, Company G gave the Advanced Technology and Planning Group within IT responsibility for standard setting and enforcement; Company H gave this responsibility to a team within its IT and Systems Development group, also housed in IT.

In contrast, some companies allocated responsibility much more to the business units. In Company E no corporate steering committee existed; instead, responsibility for standard selection and implementation fell totally to individual business units which had divisional steering groups composed of line managers. Other had committees composed totally or primarily of business executives: for example, Company A had an Information Systems Planning Committee that included the executive committee, the division presidents, and the CIO. In still others, IT made recommendations, but the top executives made the final choices.

The remaining companies created groups that drew from both IT and business units, to capitalize on the strong technical knowledge of IT and the superior business acumen of the individual business units. For example, Company C similarly had an IT standards committee whose responsibility was to scan for new technology and make recommendations to top management through the CIO. Company F's Architecture and Technology Committee included representatives from corporate IS services, data communications, and other functions, a few top-level business executives, and some business unit staff members. Company D took a different approach; they have an IT Integration Steering Group located on the business side, an IT Leadership Team, located in IT, and a vice president of Global Process Integration who oversees a Common Systems Group that acts as a steering committee focusing on remaking their business processes into global ones.

The Prototypes of the Process

The five dimensions, when considered together, form patterns that describe an organization's overall approach to the standards process. Although a large number of patterns is possible, we identified four types in our pilot study; Table 3 summarizes these prototypes.

Table 3

Prototypes for Standardization

 

Type 1

Type 2

Type 3

Type 4

Drivers of Standard Setting

technological expertise

business needs

business needs

business needs

Centralization

centralized

centralized

centralized

decentralized

Influence of IT and Business Units

IT selects and enforces; business units live within standards

IT selects and enforces; business units decide timing or can override

IT and business units jointly select and enforce

IT recommends to business units and/or support implementation; business units select standards

Decision-making Approach

directive (from IT)

negotiated

collaborative

directive (from business)/ or bridging

Oversight

IT or business oversight

mixed oversight

IT oversight

business oversight

Examples

A, H

C, F

B*,D*,G

B*, D*,E

* Companies B and D are in transition from Type 4 to Type 3

Type 1: IT Control.

This approach to standard setting emphasizes the knowledge and high level of technological expertise of the IT managers and staff. It does not ignore business needs, but interprets them within the context of the available technology. Centralizing control in IT, IT selects and enforces standards while requiring business units to live within the standards set. In line with this approach, decision-making can be characterized as directive from IT. While we would expect that oversight would support this unidirectional focus, our data suggested that two types of oversight groups actually existed. Company H, for example, used an IT and Systems Development group within IT to set and enforce standards. Company A, in contrast, included the company's executive committee, division presidents, and the CIO in their Information Systems Planning Committee which provided oversight regarding standards.

This prototype offers two major advantages. First, IT can make large, organization-wide changes because they control the standardization process. Second, economies of scale in purchasing hardware and software as a result of common standards likely reduces costs. But these advantages are also accompanied by disadvantages: users may feel that they lack any control in their computing environment. The CIO at Company A, for example, commented about the tradeoffs.

Last October we took the entire North America user community through this hoopla and it's a hard thing to do...not quite as hard as a totally new installation, but it's not much less...What we had before was the ability to make lots of little changes but we couldn't really make the big changes. And we couldn't really get enough done. Now what we have is we can make these big changes and the environment is changing too much for users.

The outcomes of Type 1 include reduced costs because of economies of scale. Increased quality can result because of the technological expertise and also due to the same scale economies. At the same time, IT must command the ability to secure the resources required to implement the standards they select. Top management oversight, either explicitly, as reflected in the composition of Company A's oversight group, or implicitly, as suggested in Company H, must support agreed-upon initiatives with sufficient resources.

Type 2: Strategic Oversight

Business needs worldwide, rather than technological expertise or individual business unit needs, drive this approach to standardization. At the same time, the strategic direction of the company provides impetus to standard setting in which IT selects and enforces standards while business units determine the timing of their implementation or can even choose to override them. Thus, the decision-making process is characterized as negotiated, rather than directive. The method of oversight reflects this involvement by IT and business units in that companies using this approach tend to have one or more committees in which both business executives and IT staff are included. Company C, for example, has an executive steering committee of all senior executives and a standards committee that scans for new technology; Company F has an Architecture and Technology Committee that includes corporate IT representatives, a few business managers, some data communications staff, and some business staff.

This standardization process facilitates the roll-out of new applications since they are distributed worldwide. Mixed oversight also empowers both IT and business units to request and implement changes in standards. Economies of scale result from the strong role IT plays. Sometimes, however, IT can lose choices when the executive oversight committee drags in setting guidelines. As the CIO in Company C noted,

I've got to get more involved in the earlier decision process. [Otherwise, by the time it gets to me] I've lost all of my choices. I mean I get to the point where I'm saying it's too late to be able to respond...And I'm telling you these battles flare up from time to time. You end up with always constantly struggling. I've got some recently with the [Division A] folks. You look at a small business that's got tremendous growth opportunity. I've met with them twice in the last month down in New York City, and they don't want to hear the corporate story.

The outcomes of this approach can include some reduction of cots through consolidation of data centers and sharing of other hardware and software. Because of the involvement of both IT and business units in standardization, this approach can also increase the quality of information systems.

Type 3: Global Partnership

Business needs also drive the global partnership approach to standardization. Type 3 organizations also are striving to create a global approach to information management, one in which information technology is seamless across units in diverse locations. IT and business units jointly select and enforce standards through collaborative, consensus-oriented decision making. Yet, both companies in this group include IT oversight to ensure that the most appropriate standards from a technological perspective are set. Company G, for example, has an Advanced Technology and Planning Group within IT for this purpose; Company D, which is moving to Type 3 from Type 4, has created an IT Leadership Team to serve this function.

This approach helps ensure that business needs within the global context are met. This makes it easier to focus on customers. Yet, it may be more expensive to the company when business unit needs predominate over global needs and reduce economies of scale in standard selection. Those companies with this prototype that truly move toward a global focus may lose the benefits of the local focus or need to create highly complex structures to ensure that both local and global benefits occur.

Type 3 organizations have experienced some cost savings, particularly on a cost per unit basis where common standards have been introduced. An increase in user satisfaction has also occurred, particularly when IT has shown that both global and local needs can be met. Company D has seen movement to this structure result in the following:

For example, we have a little program called ESL, Enterprise Software Licensing. It was a program that we did for desktop software. And we decided that we would, that corporate would go out, and get a worldwide agreement with Microsoft, Lotus, Symantec, all those guys, and we would then offer that software globally to people. We didn't tell them they had to buy it. But the financial attractiveness...so that works to get people coming into it.

Type 4: Business Unit Control

Some companies allow business units to direct the setting and enforcing of standards creating a decentralized model of standardization. In this scenario, IT recommends standards to business units, or supports their implementation, or both. Decision making tends to be directive, from the business unit. Organizations that demonstrate this prototype of standardization typically also emphasize business oversight, for example, using divisional steering committees as at Company E. Companies B and D illustrate the movement from this structure to Type 3; at Company B, the Management Committee represents the needs of the business side and at Company D, the IS Integration Steering Group represents these needs.

This approach to standard setting emphasizes the importance of local benefits in selecting standards. Company B, for example has historically operated and continues to function with strong business unit control:

What I'd say historically has happened is that these teams [within business units] have enough autonomy so that if they select something that doesn't fit well in the architecture, unless it is a grievous error they have been strong enough to drive the project through...I think we have had a culture that has been very local-centric and I think that is changing a lot."

This approach emphasizes the priority attached to the wishes of the business unit. Such responsiveness is typically accompanied by placing accountability and budgetary responsibility for IT activities within the business units. These same advantages can become disadvantages if business units lose sight of large corporate objectives, lack sufficient budget to institute appropriate standards, or fail to use the technological resources available when choosing standards. Lack of commonality across divisions may also result in duplication of effort. The satisfaction of IT staff may decrease in organizations that fit this prototype because they feel they lack control over issues that should be in the purview of IS.

One view [of the satisfaction of IT staff] is: do we have a lot of turnover? Absolutely not. We have a very low turnover. So they are happy in terms of morale, but as far as being able to deliver I'd say they are all pretty frustrated. I think a lack of having some real intervention-oriented strategy, although they would resist it, they still want it. They want to know what the tools are they should be using and they want to have confidence that the work that has been done to make that choice has taken their interests into account and that the choice has been made in a thoughtful way.

This prototype often results in an increase in costs, but simultaneously an increase in services to division. Because the divisions determine the IT budget as well as standards, some tradeoffs likely occur. In general, costs tend to increase because of the duplication of effort. Because this approach reinforces a strong business orientation and thereby facilitates organizations responding rapidly to changes by competitors, little commonality across business units may exist. This may result in constantly changing the IT skill set, resulting in frustration among IT staff members.

Implications and Conclusions

Understanding these prototypes, the context in which they would work best, their advantages and disadvantages, and their outcomes is important for both IS executives and top management of organizations. Our interviews suggest that overall the approaches used to set and maintain standards were quite varied. Nevertheless, the particular approach used by individual organizations seemed to be effective. This diversity suggests that organizations can manage the standardization process best by understanding what needs and knowledge drives standard setting in their organization and developing roles and methods of oversight to match them. For example, organizations that value knowledge of technology over needs of individual business units likely will establish a stronger role for IT than those who emphasize the primacy of the business units. Similarly, those organizations for whom business unit adaptability, responsiveness, and autonomy is key, will de-emphasize the control exerted by IT in standard setting and maintenance.

At the same time, our examination of prototypes suggests that the roles, approaches, and method of oversight selected are not necessarily mutually reinforcing. Instead, some organizations chose to ensure the voice of both IT and business units by creating potentially conflicting bases of control.

Finally, some prototypes likely work better as part of specific structures for delivering IT services. Clearly, an IT control standardization process complements a highly centralized, functional structure; a business unit control process complements a more decentralized structure. Yet some argue that an IT control process works well for decentralized, organic organizations because it brings control to these highly flexible, highly adaptable, innovative structural forms. Choice of a standardization process model ultimately must occur within the context of the organization's goals, culture, business needs, and technological expertise.

We have identified and examined four prototypes. More types, with differing combinations of the five dimensions, likely exist. Future research needs to validate these four types and determine whether additional prototypes exist. In addition, future research needs to examine which prototypes fit best with which organizational structures.

Finally, it is clear that organizations change their approach to standardization, as well as other IT activities, over time. Both Companies B and D were changing the process, with different degrees of speed and success. Movement between prototypes occurs and is to be expected. Organization leaders should periodically reassess the effectiveness of the prototype for standardization chosen as market forces, technology, and business needs change.

References

Adhikari, R. (June, 1996). Taking care of business. Software Magazine, 57-62.

Alavi, M. and Young, G. (1992). Information technology in an international enterprise: An organizing framework, in Global Issues of Information Technology Management, Palvia, S.C., Palvia, P.C. and Zigli, R.M. (eds), Harrisburg, PA: Idea Group Publishing, 495-516.

Allen, B. and Boynton, A. (1991). Information architecture: In search of efficient flexibility. MIS Quarterly 15(4), 435-442.

Alter, A.E. (February 26, 1996). Change in the weather. Computerworld, 77,79.

Bartlett , C. and Ghoshal, S. (1989). Managing Across Borders: The Transnational Solution. (Boston: Harvard Business School Press).

Cale, E.G., Jr., Kanter, J., and Saia, P.J., Jr. (1993). The changing locus of IT decision-making: The evolving role of the business manager, CIMS Working Paper Series 93-07, Center for Information Management Studies, Babson College.

Cash, J.L., Jr., McFarlan, F.W., McKenney, J.L. & Applegate, L.M. (1992). Corporate Information Systems Management: Text and Cases, 3rd ed. (Homewood, IL: Irwin.

Daniele, E. (1994). Conquering the culture gap. Insurance & Technology 19(3), 20.

Dearden, J. (1987). The withering away of the IS organization. Sloan Management Review 28(4), 87-91.

Galbraith, J. R., Lawler, E. E. III, and Associates. (1993). Organizing for the Future: The New Logic of Managing Complex Organizations. (San Francisco: Jossey-Bass).

Gibson, R. (1994). Global information technology architectures. Journal of Global Information Management 2(1), 28-38.

Gordon, S. (1993). Standardization of information systems and technology at multinational companies. Journal of Global Information Management, 1(3), 1-10.

Jarvenpaa, S.L. and Ives, B. (1993). Organizing for global competition: The fit of information technology, Decision Sciences 24(3), 547-574.

Kelly, T. (September 15, 1991). A wise consistency. CIO, 22,24.

Kim, K.K. (1988). Task characteristics, decentralization, and the success of hospital information systems. Information & Management 19(2), 83-93.

LaPlante, A. (February 20, 1995). Life in the foreign lane. Computerworld, 117-121.

Mazur, L. (August 11, 1994). The new invasion of Europe; How companies make the local/central choice. Marketing, 18-20.

Meek, B.L. (1990a). Changing people's attitudes: Personal views. Computer Standards & Interfaces 10, 29-36.

Meek, B.L. (1990b). Problems of software standardisation. Computer Standards & Interfaces, 10, 39-43.

Miles, M.B. and Huberman, A.M. (1994). Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed, Thousand Oaks, CA: Sage Publications.

Nohria, N. and Eccles, R. G., Eds. (1992). Networks and Organizations: Structure, Form, and Action. (Boston: Harvard Business School).

Palframan, D. (December, 1991). The great software sort-out. Management Today, 86.

Pastore, R. (November 15, 1995). Dutch masters. CIO 9(4), 58-61.

Perry, W.E. (1992). Quality concerns in software development: The challenge is consistency. Information Systems Management 9(2), 48-52.

Rosenthal. D. and Bakun, M. (1992). Software application standardization for multi-national corporations. Proceedings of the Northeast Decision Sciences Institute, Cambridge, MA.

Rosenthal, D. and Gordon, S. (1993). Standardization of information systems and technology -- report on data collection. Unpublished Working Paper.

Strauss, L.L. and Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques, Newbury Park, CA: Sage Publications.

Sweeney, J. (October 2, 1995). IBM prepares to offer global services under common name. Computer Reseller News, 6.

Von Simson, E.M. (1990). The 'centrally decentralized' IS organization. Harvard Business Review 68(4), 158-162.

Westoby, R. (1996). Combining forces. Financial Executive 12(3), 28-30.

 

Footnotes

[1] Gordon, 1993.

[2] Kelly, 1991; and Palframan, 1991.

[3] Cash et al., 1992; Meek, 1990a; Meek, 1990b; Perry, 1992; Rosenthal & Bakun, 1992; and Rosenthal and Gordon, 1993.

[4] Dearden, 1987; Kim, 1988; Von Simson, 1990; Alter, 1996; and Adhikari, 1996.

[5] Pastore, 1995; and Alter, 1996.

[6] See Galbraith, Lawler, and Associates, 1993; and Nohria and Eccles, 1992, for a description and analysis of these new organizational forms.

[7] Daniele, 1994; and LaPlante, 1995.

[8] Bartlett and Ghoshal, 1989.

[9] Jarvenpaa and Ives, 1993.

[10] Mazur, 1994; Sweeney, 1995; and Westoby, 1996, address the increasing centrality of the business unit in the analysis of corporate structure. Allen and Boynton, 1991; and Cale, Kanter, and Saia, 1993, address the role of the business unit in the context of delivering IS services.

[11] This methodology is suggested by Miles and Huberman, 1994; and Strauss and Corbin, 1990.


This file was last updated by Steven Gordon on February 18, 1998.