Executive Information Systems without Executives

Executive information systems (EIS) are the most-recent computer-based information systems to have emerged with the intention of providing executives with the information they require to run their businesses. Some advocates of these systems see them as a panaceathe long-awaited moment when computing will provide meaningful and significant assistance to top management. However, others have questioned the extent to which these systems are suited to executive work, and see them as yet another passing fad in a field that has had more than its share of the same. This paper discusses some of the potential benefits to executives that have been identified in the literature as arising from the use of EIS. The results of a field study which investigated the extent to which EIS are actually used by executives are reported. The findings corroborate the now widespread finding that executives are not the main users of EIS. Some reasons for this are identified and possible remedies are suggested. 1. The Emergence of EIS and the Potential Benefits of EIS for Executives While there are many differing definitions of EIS, there is broad agreement on the characteristics of EIS. They are easy to use, invariably making use of a graphical user interface; they can capture both external and internal information of relevance to executives (Watson, Rainer and Koh, 1991); they can also cater for soft, non-quantitative information, and can provide highly-aggregated information, while still allowing selective 'drill-down' to the underlying detail if required. Trends can be monitored and represented graphically, and various office support functions such as electronic mail, diary and calendaring facilities are often provided (Rockart and De Long, 1988). The concept of providing information to executive management is not something newonly its computerisation is new. Given that top management have up to now succeeded in acquiring the information they need to run their business without direct access to computerbased information systems, the question arises as to why EIS are achieving prominence. The literature suggests a number of broad factors as relevant: firstly, there is a ‘pull’ factor in that executives are suggested to be more computer-literate and willing to become direct users of computer systems (cf. Houdeshel and Watson, 1987; Paller and Laska, 1990; Rockart and Crescenzi, 1984; Rockart and De Long, 1988; Rockart and Treacy, 1982; Watson et al., 1991). For example, one survey suggests that more than twenty percent of senior executives have computers on their desks (Paller and Laska, 1990), although there are wide variations in the estimates as to how many executives are direct users of computersfrom as low as one percent of executives (Rae, 1986) to a figure of ten percent (Rockart and De Long, 1988). A number of arguments have been put forward to explain this increased use of computers at executive level. For example, it has been suggested that middle managers who have been making direct use of computers in their daily work are being promoted to executive level (Paller and Laska, 1990). This new breed of executives do not exhibit the fear of computer technology that has characterised executive management up to now and are quite willing to be direct users of computer technology. Also, executives have heretofore managed to remain relatively untouched by the computer, seeing it as a tool appropriate to others of lower staturea position reinforced by some influential researchers (Dearden, 1983; Salerno, 1985). However, researchers have argued that there may be feelings of guilt among executives, due to having missed "the wave of end user computing that has swept through the rest of their organisations" (Harvey and Meiklejohn, 1989 p.124). Complementing this suggestion of executive demand for EIS, a number of ‘push’ factors have been identified as motivating the development of EIS, most notably the highly competitive nature of today's business environment which requires executives to act more quickly and effectively if an organisation is to remain competitive (cf. Burkan, 1988; Friend, 1991; Fireworker and Zirkel, 1990; Rockart and De Long, 1988; Rockart and Treacy, 1982; Watson et al., 1991; Wetherbe, 1991). Another ‘push’ factor is the fact that technology has evolved to the stage where computer-based support for top management may be feasible (cf. Paller and Laska, 1990; Rockart and De Long, 1988; Rockart and Treacy, 1982; Whymark, 1991). These technological advances have seen the replacement of character-based user interfaces with "user-seductive" graphical interfaces that allow painless access to computerbased information. The literature also identifies a number of significant benefits which EIS could provide for executives, including easing of information overload; improved efficiency in certain aspects of their work; increased span of control; direct access to information rather than having to rely on intermediaries; and improved mental models of the business. These are briefly summarised here. 1.1 Easing Information Overload Several researchers have identified the information overload problem as one of the most serious problems confronting top management today (Kotter, 1982; Mintzberg, 1975; Naisbitt, 1982; Zuboff, 1988). The executive may be monitoring several projects, and for each one has to track a vast amount of information because it is not possible to predict where problems are going to occur. This leads him to play safe and request all the information he can get, but this bewildering choice and quantity of information may cause him to miss the wheat for the chaff (Mintzberg, 1975). Drucker (1988) suggests that many highlycomputerised companies continue to operate under the simplistic belief that more data equals more information. Thus, executive management, despite being already inundated with data, exhibit an insatiable appetite for more. Compounding the problem of the executive's need to monitor such a vast amount of information is the fact that so much of the information generated is of little benefit in helping to run the organisation. Indeed, a US survey reports that 96 per cent of pages in management reports are never read (Harvey & Meiklejohn, 1989). Also, Mintzberg (1975), in his seminal empirical study of the nature of executive work, concludes that most computer-generated reports are just skimmed ritualistically, if read at all. Researchers have estimated that an executive may spend up to half his time getting the information he requires (Wetherbe, 1991). However, it has been argued that EIS can help greatly by streamlining this process, highlighting essential information and discarding the irrelevant. Problems of timeliness and relevance which have plagued traditional reporting systems can be addressed (Harvey and Meiklejohn, 1989). 1.2 Improved Efficiency Much of the work of executivestheir use of intuition and the interpersonal and verbal aspects of their workcannot be well-supported by computer systems. However, while not underplaying the importance of these aspects of executive work, researchers have made a case for executive use of computer support in attainment of corporate goals. Isenberg (1984), for example, suggests that if managers fully trusted their intuition, there wouldn't be any need for rigorous and systematic analysis, but, in practice, they try to achieve a match between "gut" and "head". Rockart and De Long (1988) use Isenberg's findings to support their argument that office support applications be included as part of EIS. Applications such as calendaring facilities, diary facilities, electronic mail, for example, have the potential to make the executive more efficient, rationalising the aspects of executive work that lend themselves to computer support, thus freeing up the executive to spend more time on complex unstructured tasks, which is more properly the remit of the executive. 1.3 Increased Span of Control Researchers have reported the need to increase the span of control in modern organisations. Drucker, for example, calls for "skimming management's midriff", stating that middle management is "overstaffed to the point of obesity". These extra layers of management cause rigidity and inertia whereas flexibility and responsiveness are needed. However, the electronic mail capability of EIS can contribute significantly to increasing the span of control, allowing the executive to keep in close contact with subordinates. Also, EIS can help to motivate and focus the organisation towards top management's goals and, because executives are perceived to be more readily able to access this information through the use of EIS, it extends their psychological presence throughout the organisation (Rockart and De Long, 1988). 1.4 Direct Access to Information Traditionally, executives have relied on intermediaries to perform the information gathering function for them. However, this practice has a number of negative implications. Mintzberg (1975) refers to what he terms the "dilemma of delegation". The executive is the best person to scan the information since he has all the intuition locked in his brain, but he has to leverage his scarce time and must delegate some information scanning responsibilities to his subordinates who may have less well-developed business intuition. A similar argument has also been made by El Sawy (1985). The basic problem is that subordinates filter out unnecessary information, but this is a subjective process as subordinates will shape the data according to their ideals and interests. At best, this filtering causes the unintentional loss of information that might be valuable if available to the executive. At worst, executives are presented with a sanitised version of the information, as subordinates, unwilling to allow greater visibility into their areas of responsibility, suppress important information. Thus, there may be great advantages in the executive "dirtying his hands in the data" (Rockart and Treacy, 1982), playing with the information directly and perhaps uncovering valuable business information. Other researchers have stressed the importance of this "playfulness" in exploring possibilities (March, 1988), thus increasing the opportunity for the serendipitous discovery of new and important information about the business. 1.5 Improved Mental Models Rockart and De Long define a mental model as a "cognitive construct that describes a person's understanding of a particular segment of the managerial world" and state that the enhancement of executive mental models is one of the areas in which EIS have the greatest potential. They identify a number of ways in which EIS can improve an executive's mental model, including improved access to external data; the ability to combine data from multiple sources and present it in more meaningful formats; improved analytic and modelling capabilities which allow assumptions to be surfaced and tested. These improved mental models may afford the executive fresh insights into how he conceptualises and understands the business. 2. Research Study The factors identified in the previous section to explain the emergence of EIS, and the significant benefits that these systems are argued as providing to executives, represent plausible arguments in favour of EIS. However, given the fact that the nature of executive work is not well-understood (Rockart and De Long, 1988; Zuboff, 1988), there is the distinct possibility that EIS represent a solution being proposed without adequate analysis of the problem. In the past, very little research in the EIS area has been of an empirical nature (Fitzgerald, 1991). This research study was thus concerned with investigating the factors behind the decision to implement EIS, and the manner in which EIS were being used in practice in a number of organisations. A case study approach was adopted as this provides richer detail than that possible through the use of a survey, for example. This has been argued to be especially relevant in the case of EIS (McBride, 1997). Four organisations were chosen for the study, drawn from the food processing, finance and state sectors. All had developed some EIS systems with the average length of time since EIS initiation just under two years. The organisations were chosen on the basis of being able to get sufficient access across all organisational levels, from the level of chiefexecutive/managing director through all other levels and functional areas. While this criterion for organisational selection may be seen as a weakness of the research methodology, it is worth noting that all were large companiesannual turnovers ranging from £200m. to £700m., and numbers employed ranging from 600 to 4,000 employees. Also, it was felt that interviewing several personnel in each organisation (average of 5 interviewees per organisation) was important in that it was leveraging depth and ensured that several perspectives were considered. A total of 21 people were interviewed, with interviewees falling into the two broad categories of EIS developers and EIS users (executives and middle management). While not wishing to adopt a rigid format for the interviews, questionnaires were used, both to act as an aide memoire and to give some structure to the interview process, with separate questionnaires for the EIS users and EIS developers. 3. Research Findings The findings of the study in relation to the factors involved in the decision to implement EIS are discussed in detail in Fitzgerald and Murphy (1994). Briefly, summarising here (see Table 1), the findings do not support the conventional literature argument that EIS are being introduced because computer-literate executives, driven by the competitive nature of the prevailing business environment, are requesting these executive-friendly computerised information systems. Rather, the argument that fits better with the findings of this study could be cast as follows: Technological advances have made available computerised systems which overcome the traditional problems associated with the syntax-amnesic executive's reluctance to use computers. In addition to this, the literature contains many evangelical examples of companies who have successfully implemented EIS. This has caused an EIS envy (apologies to Freud) phenomenon whereby IS departments, outside the strategic epicentre in many organisations, see EIS as a potential means of winning back influence at top management level. A similar view of the multiple layers of factors influencing EIS development has been reported by Nandhakumar (1996). Table 1 Assessment of Relevancy of Factors in the Introduction of EIS Number of Number of Factor Interviewees Organisations Availability of appropriate enabling technology 15 4 IS department push 10 3 Competitive nature of the business environment 4 2 Increased readiness of executives to use computers 1 1 However, executives are also supporting EIS development, principally because they are unsure of the role these systems might play, and fear competitors might make use of them and achieve competitive advantage. However, such motives for introducing EIS do not augur well for the committed use of EIS by executives. Indeed, the findings of this study show that the majority (68 per cent) of users of EIS were at middle management level (see Table 2). Some of the reasons for the low level of executive usage are discussed in detail below. Table 2 Users of EIS by Organisational Level Avg. per Total: All Company Minimum Maximum Companies %age Executive Management 5 2 10 18 32% Middle Management 10 3 23 39 68% 3.1 Simplistic View of Executives and their Information Requirements The principal instigators of the move to introduce EIS were the IS department in three of the organisations studied (see Table 1). However, the IS department were not able to gain ready access to executive management. Thus, they were forced to collaborate with middle management on EIS developmenta factor which had implications in relation to the information content of EIS. (This issue is discussed in a subsequent section). However, if EIS are going to be used by executives, it is clearly reasonable to expect that they would be quite involved in the process of identifying information requirementsthe rationale being that the executive is the best person to determine the information content of EIS as he can thus identify the content and format of the information he considers to be most important. This helps ensure consistency and standardisation in the information presented, and it also sends strong signals to other levels of the organisation as to what the executive considers to be important. This study found very little executive involvement in the identification of information requirements for the systems being developed. In two organisations, EIS developers had used largely their own initiative as to the information the systems should contain. In another organisation, the information content of the EIS was determined by replicating the existing hard-copy reports received by executives. Only one organisation had carried out a formal procedure (identification of critical success factors (CSFs) and key performance indicators (KPIs) to determine executive information requirements (see Table 3). Table 3 Identification of Executive Information Requirements Number of Method Organisations EIS director used his own initiative 2 Replication of existing hard-copy reports 1 received by executives Formal process based on critical success 1 factors and key performance indicators Given the low level of actual contact between executives and EIS developers, it is not surprising that in some cases, EIS developers seemed to make the assumption that data and information were synonymous, and a simplistic view of executive information requirements was taken, with EIS being viewed as akin to a receptacle into which information could simply be poured on the basis of the extent to which the information lent itself to capture, rather than because it was actually requested by executives. Strong support for this interpretation comes from the opinions expressed by several executives as to the unsatisfactory information content of their EIS systems. For example, one executive suggested that his company’s EIS contained a “ragbag of information”. In another company, an executive distinguished between “information that is nice to know, and information that one needs to know”, suggesting that the information in his company’s EIS systems was of the former type. Even though IS personnel tended to an overly-simplistic view of executive information requirements, they, nevertheless, do have a role to play in EIS development. They have the requisite technical skills and have the experience of mistakes made on system development projects in the past. For example, this study found that EIS developers who did not have an IS background had great expectations that EIS would eventually provide real-time 'drill-down' into operational systems. However, the EIS developers with an IS background were wellaware of the practical problems that could arise with such a strategy. 3.2 EIS Viewed as Graphical MIS Executives spend their careers trying to construct and maintain their information system, building up an extensive interpersonal network (Isenberg, 1984). However, as already stated, EIS developers were typically of the view that executive information requirements are easily understood and codified into a computer-based information system. EIS were thus viewed more as MIS-type systems with a graphical interface, rather than a new and distinct category of information systems in their own right. To this end, the systems as they had been implemented were providing rapid access to pre-defined categories of information which was generally concerned with historical financial performance. EIS systems were thus very tightly-coupled to existing MIS systemsa strategy which is not well-suited to the needs of executives (El Sawy, 1985). Furthermore, Turban and Schaeffer (1989) have suggested that EIS are quite different to MIS, and as such, little will be transferable from existing MIS systems. Indeed, a different development methodology is recommended (Volonino and Watson, 1990). Notwithstanding this, in two of the organisations, EIS developers stated that the same approach to developing MIS should also be applied when developing EIS. 3.3 Information Content of EIS Researchers have argued that the use of EIS as a control mechanism is a sub-optimal one, in that EIS should play a role in strategic planning (Friend, 1989; Rockart & De Long, 1988). Similarly, researchers have emphasised that the information focus of EIS should not be primarily on financial figures but on more market-oriented soft and external information (Bentley, 1989; Rockart & De Long, 1988; Shoebridge, 1988; Volonino & Watson, 1990; Watson et al., 1991). However, this study found that all organisations were using EIS primarily as a control mechanism. In fact, this was seen as the most useful role for EIS (see Table 4). This strong emphasis on control is probably to a large extent a reflection of the fact that the primary users of EIS were middle managers rather than executives, operational control being the primary responsibility of the former. Also, the information content of EIS was predominantly financial, with little in the way of external or soft informationsimple daily stock market share prices were typically the extent to which external information was provided. The perception of EIS as a control mechanism may also be due in part to this focus on financiallyoriented historical information. There are negative aspects, however, to concentrating on financial performance figures in that it may be too narrow a focus for overall organisational success, which is the primary concern of executives. By concentrating on historical financial performance, EIS may be criticised for fulfilling the sub-optimal role of ‘coming in after the battle is over and counting the wounded’, rather than playing a more useful role in providing information to support the strategic direction set for the organisation. If EIS could provide such information support, the take-up by executives would be much greater. Table 4 Rating of EIS as a control mechanism Average Lowest Highest Interview Question Value Value Cited Value Cited "The most useful role for EIS is that of control mechanism" 4.7 4 5 1. Strongly disagree. 5. Strongly agree 4. Reasons Behind Executive Support for EIS: The Hidden Agenda Although executives were not leading the push to develop EIS, and did not rate EIS as having any major benefits for themselves (see Table 5), there was some support for EIS from several executives. However, the motives underpinning this support are quite interesting, in that they were primarily of a political nature, rather than because EIS were perceived as having any direct benefit to executives. Table 5 Executive rating of EIS in terms of benefits provided Average Lowest Highest Dimension Value Value Cited Value Cited Benefits Provided 3.2 1 4 1. Little Benefit. 5. Major Benefit 4.1 EIS Anxiety: A Defensive Strategy The EIS envy term was introduced above to explain why the IS department were so much in favour of the high profile nature of EIS development. However, a similar phenomenon occurred at the executive level where it manifested itself in terms of EIS anxiety. For example, several senior executives, very much aware of the current interest in using information technology for achieving competitive advantage, expressed their concerns that their competitors might make use of EIS to gain competitive advantage, and they feared they could be at a disadvantage by not having some EIS development underway. These executives, while conceding that they would not become users of EIS, were nevertheless strongly supportive of EIS development in their organisations. This finding is consistent with the argument stated earlier, namely that executives may be supporting EIS due to feelings of guilt at having missed out on the end-user computing phenomenon in the past. For other, less-senior, executives, internal organisational politics was very definitely a factor in their decision to become involved in EIS. These executives sought access to EIS through fear that they might be marginalised if other executives had access and they didn't. However, because it was seen as a hedging exercise, they were not prepared to commit more than the minimum necessary involvementenough to ensure that if EIS became a vital component in the executive armoury, they would be sufficiently informed to stay in touch. A 'Catch-22' scenario emerges in that executive time is very precious, and they are not likely to expend it on EIS development, unless they believe such a system could truly benefit them. However, if they do not commit the time, any systems developed will have serious shortcomings. 4.2 Business Process Redesign While 70 percent of executives interviewed did not perceive EIS as being useful in supporting their work, in one organisation, top management were very supportive of EIS development. This organisation had appointed senior executives to the executive and operating sponsor roles (the only organisation in the study to formally do so). However, it emerged that executives in this organisation had a hidden agenda, seeing EIS in terms of the opportunity it represented for facilitating business process redesign. The organisation in question had recently undergone privatisation and were actively involved in reengineering business processes. Several researchers have identified the thorough examination and understanding of existing business processes as an important preliminary stage in business process redesign (Bevilacqua and Thornhill, 1992; Davenport and Short, 1990; Kitch, 1992; Nedzel, 1992). However, the organisation in question had been experiencing major difficulties in getting managers to commit time to examining existing business processes and management practices. Senior executives saw EIS as a possible means by which management attention could be focused on these issues. Thus, they were using EIS development to facilitate this, and saw the fundamental examination of business processes that was motivated by EIS development as a major benefit. Executives in this organisation, including the executive and operating sponsors, were not overly-concerned as to whether any EIS systems would emerge and survive as operational systems. For these executives, the eventual emergence of operational EIS systems was a bonus, in that the process of critically examining management processes was rated as far more valuable. To most advocates, this role of EIS as a means to a business process redesign end may seem to be akin to the tail wagging the dog. Ironically, one of the objectives of EIS is the institutionalisation of the corporate knowledge-basea goal common to business process redesign as well (Kitch, 1992). Ironically, the EIS systems that eventually emerged in this organisation were rated higher by executives in this organisation than in any other. Also, this organisation had many more EIS users, notably at executive level, than any other. This is perhaps indicative of the fact that, rather than the EIS director using his own initiative as to the information the EIS should contain, the information content of their EIS had a sound underpinning in actual management processes. EIS developers attended formal board meetings and other executive planning sessions. They also used EIS itself to track the manner in which executives were actually using the system. Thus, the developers gained a very good insight into executive work practices, and were able to incorporate this into their EIS systems. Researchers have suggested that the effectiveness of the process used to introduce technology strongly influences its ultimate impact (Rockart and Short, 1988). Certainly, in this organisation, the process followed led to better likelihood of defining the right problem, and thus, developers had a better chance of providing the right solution. 5. Conclusion Executives rely primarily on trusted support staff to provide them with the information they need to run their business, and the findings of this study indicate that this is not going to change. Nevertheless, the notion that EIS are not of any benefit to an organisation if they do not become operational or if they are not used by executives is open to question. EIS development may have fulfilled a purpose by facilitating organisational change, such as in the case of business process redesign reported in this study. Thus, EIS development is much more an organisational initiative than a technological initiative, and as Paller and Laska (1990) suggested, the journey may be more important than the destination. However, in most organisations where EIS are introduced, it is presumably with the intention, initially at any rate, that they be used by executives. While it was too early to say how all the systems that had been developed in these organisations would eventually fare, the evidence to date suggests that many of the systems have failed to elicit even an initial stage of euphoria among executive ranks, and were not likely to be used by executives in any meaningful day-to-day situation. Certainly, the lack of real executive involvement might indicate impending failure in some organisations. The paradox emerges that executives won't use EIS until it becomes part of the management process, and this will not happen until executives commit adequate time to the process of developing EIS. The introduction of EIS should represent a deliberate business-driven intervention rather than accidental or serendipitous change. Yet, for many of the executives in this study, it represented a project that might reap some dividend, but it was not being allowed to consume much of their time or attention. A corollary of this is that executives were not willing to participate meaningfully in the identification of information content of EIS. In the ensuing vacuum, developers took the initiative as to what should information should be provided. However, as already stated, there was a tendency towards over-inclusion, as information was gathered which lent itself to ready capture. In fact, such a simplistic approach to the determination of the information content of EIS may do more harm than good by diluting the perceptions of executives as to what the EIS concept represents in terms of the commitment required to achieve results. In short, if EIS are to be used by executives, then the latter must take a much more proactive role in the development of these systems, rather than allowing the IS department to develop EIS in isolation. The result of the latter all-too-common scenario is that the IS department will consult with middle management with whom they can gain worthwhile access, or will focus on financial control information as most IS departments report through the financial controller function. Both of these options lead to rather sterile, historical performance type information being provided. EIS development thus demands substantial and proactive executive commitment if it is to have any chance of success. There is no royal road to EIS, and relying on computer packages alone is not enoughthey cannot produce the desired effect if not firmly founded on executive processes in the first place. As Bentley (1989) rather pessimistically concludes: The real-world executive information system is a complex, many-levelled one with both formal and informal information inputs. It also includes all the executive’s experience and training and personal knowledge. To suggest this can all be placed onto a computer system is, to say the least, ill-informed.