While figures vary over time and across industries, the fact is that for most firms, information technology (IT) investments constitute their largest capital-spending item. On average, large US firms spend $300-500 million/year (or 3-4% of total revenue) on IT, with $50-90 million of those dollars invested in new IT products and services.8 Nevertheless, industrywide managerial attitudes regarding the value of IT innovation have fluctuated over the years. An emerging pattern is characterized by periods of optimism, in which IT innovation is celebrated as a panacea for all business ills, followed by periods of pessimism, in which doubt prevails about the value of investing in new IT, with persisting arguments regarding the ease with which IT can be replicated (Figure 1).
In all fairness, in the modern hypercompetitive world it is unlikely that any single investment in IT (or non-IT, for that matter) will lead to a sustained competitive advantage. Instead, what does appear to make a difference is a company's ability to innovate with IT over time. Wal-Mart is a case in point. As Friedman3 notes, "Wal-Mart ... was the first to computerize, the first to use wireless, the first to really deploy RFID ... they adopted and adapted faster to new technology than any other retailer in the world. And you've got to give them credit for that. You've got to worry about and be troubled by some of the brutal side of their business practices. But at the end of the day ... [they] ... out-innovated all their competitors." Similar stories can be found elsewhere in the business world: Harrah's in the entertainment industry, Equifax in credit reporting, RR Donnelley in printing, and Harley-Davidson in the motorcycle industry. What these companies demonstrateand many innovative IT-adopters corroborateis that competitors have a hard time imitating and keeping up when a series of IT investments have become integrated with procedural and organizational innovations over the course of several years. In other words, while a single investment in new IT might be easy to copy, it is much more difficult for competitors to replicate a company's ability to innovate with IT over the longer term. This is important for managers to note because those capabilities that are valuable and not easily replicated are more likely to be a source of competitive advantage.
To date, only theoretical research has been conducted in this area of IT innovation.11 In this article, we present empirical results that support the belief that the ability to innovate with IT over time is not easily replicated by competing firms. Given the vast amounts of money currently spent on new IT and, consequently, the high stakes involved in IT innovation initiatives, such evidence is critical. In this article we address the following questions: How likely is it that a firm that has out-innovated its competitors this year will be able to repeat this performance in the following year? In other words, is IT innovation persistent? How do fluctuations in industry-wide managerial attitudes towards IT affect the persistence of IT innovation? Are innovative firms more likely to out-innovate their competitors during periods of managerial optimism or pessimism? How likely is it that a firm will go from a state of non-innovation to being able to out-innovate its competitors within a relatively short period (3-4 years)? In other words, how long does it take for a firm to acquire and develop the ability to out-innovate its competitors?
We obtained data for our study from two sources: InformationWeek and Hoover's Handbook of American Business, from 1997 to 2004. This period was chosen because, as Figure 1 shows, it contains the approximate boundaries of a complete cycle in managers' perception regarding innovation with IT. InformationWeek publishes an annual list of the largest US companies ranked by their innovative use of IT. Based on a detailed survey of IT executives, the publication determines the amount, type, and use of IT investments for each company. The companies are evaluated in terms of their business-technology strategies and deployment of investments in IT architecture, infrastructure, business, and e-business application. Incidentally, IT budgets are not a deciding factor in the rankings. The end result is an annual list of 500 firms that are classified as IT innovators because they have demonstrated a "consistent pattern of technological, procedural, and organizational innovation."
Our first step in the study was to produce a dataset of 1187 firms that have attained the IT innovator status at least once during the years 1997-2004. Since the InformationWeek list omits non-innovative firms, however, we turned our attention to one of the main criticisms that has been raised regarding IT innovation: competitors can easily replicate it.1 Using Hoover's Handbook, our second step was to identify up to three top competitors of comparable size for each one of the IT innovative firms in our first dataset. Limiting our selection to direct competitors, we were able to control for size (revenue) and industry structure (SIC classification).
This second step increased the number of firms in our dataset to 2211. It also revealed two facts: first, many competitors of the IT innovative firms were already included in the original list of 1187 IT innovative firms. Second, among the group of top competitors there were 1024 firms that did not appear in the list of IT innovators during this period. These trends are consistent with previous studies, which have argued that the use of IT varies greatly among industry sectors. For some firms and their competitors, competing through new IT is the "name of the game," whereas for others, the adoption of new IT is less of a strategic need.9 In broad terms, companies in certain industries, for example, primary and construction, have a relatively low need for new IT. For companies in industries such as banking and financial services, however, the need to innovate and compete with IT is typically much higher. Table 1 shows that our final dataset, derived from the two sources, covers firms from a wide spectrum of industries.
A review of the business-technology practices of firms from our list of IT innovators indicated that numerous firms have out-innovated their competitors. Firms such as Harrah's, Waste Management, Capital One, and RR Donnelley, approach IT innovation in relation to firm-specific idiosyncrasies of internal processes, products, and markets. Prior theory has argued that such organizational 'mindfulness' in IT innovation is embedded in the management training of individual members, and becomes part of the company's culture.5,11 This means that a firm's ability to innovate with IT is diffused across people, processes, and products, and is based on decisions and actions taken in prior years. Hence, IT innovation is persistent.
Our prediction, therefore, was that among competing firms (same industry and similar size) the probability that an IT innovative firm will be innovative also in the following year is much higher than the probability that a non-IT innovative firm will become innovative. To test our prediction we classified firms in the dataset as innovative and non-innovative. For each year from 1997-2004, a firm was classified as IT innovative if it had appeared in the list of IT innovators for that year. Based on this classification, we calculated transitional probabilities, that is, the probability that a firm would maintain its current status (IT innovative or non-innovative) in the following year. The results, summarized in Panel A of Table 2, strongly supported our prediction. The probability that a firm will remain IT innovative is relatively high (68%), while the probability that a firm will escape from its non-innovative status is small (9%).
Within the group of large US firms, there are both IT innovative and non-innovative firms. The latter may not compete this way for one of two reasons: because they don't think that it is necessary to innovate with IT or because they cannot innovate with IT. In any case, if a firm has out-innovated its competitors this year, it is more likely that it will be able to repeat this next year. Thus, IT innovation is persistent. The implications of this finding are very important for managers. If IT innovation is persistent; some firms have a unique capability.
New technologies tend to go though a "Hype Cycle."4 The cycle starts with a technological breakthrough that generates significant publicity that is accompanied by over-enthusiasm, and often-unrealistic expectations. During this period, managers' perception towards IT rises, and companies with or without IT innovation experience invest in new IT projects with little or no concern for the IT budget. Rising IT budgets and a swelling of the ranks of firms that try to innovate with IT are the manifestations of this over-enthusiasm. This opportunistic approach to IT innovation has been dubbed as a mindless, follow-the-pack, and computing-by-fad approach. Some firms that invest with this mindset may be successful; however, most will fail.4 These firms, called opportunistic innovators, are not likely to out-innovate their competitors in the future. Therefore, we predict that persistence of IT innovation will be lower during periods of over-enthusiasm.
This failure to meet unrealistic expectations leads the Hype Cycle into a 'trough of disillusionment.'4 Opportunistic innovators abandon ship, and managers' confidence in IT innovation's promises falls. Shrinking IT budgets and a scaling back of new IT projects generally accompany this mass exit.6 Nevertheless, a smaller group of firms continue experimenting with the new technology and gradually understand its firm specific benefits and practical applications. These firms, called systematic innovators, are more likely to out-innovate their competitors in the following years. Therefore, we predict that persistence of IT innovation will be higher during periods of IT disillusionment.
The dot-com boom and crash was a complex phenomenon driven by investor enthusiasm over new technology, Y2K, globalization, and numerous other factors. During this period, managers' perception towards IT closely mirrored the economic cycle: their perception went from over-enthusiasm during the dot-com boom to harsh disillusionment after the crash (Figure 1). This full cycle in managers' perception offers a unique opportunity to test our second question: "How do fluctuations in industry-wide managerial attitudes towards IT affect the persistence of IT innovation?" We split the dataset into a pre-Y2K (1997-2000) and post-Y2K (2001-2004) period, and, as in the previous question, we calculated the transitional probabilities in each sub-period. The results, summarized in Panel B of Table 2, support our prediction. By contrasting the pre- and post- Y2K periods, we demonstrated the probability that a firm would out-innovate its competitors increased from 60% in the pre-Y2K period of over-enthusiasm and rising IT budgets to 74% in the post-Y2K period of disillusionment.
The behavior during the pre-Y2K period can be explained by efforts to fix the Y2K problem and by the dotcom bubble. The former introduced a sense of incoming doom, which made rising IT budgets possible. The latter introduced a sense of euphoria and optimism around everything that had to do with dot-com and e-commerce. The result was a bandwagon type of approach to IT innovation. The collapse of the dot-com bubble, coupled with the feeling that the Y2K scare had been exaggerated, led to subsequent disillusionment of management with IT. Therefore, at least within the group of the large U.S. firms included in our study, there seem to be two types of IT innovative firms: systematic and opportunistic. During periods of optimism the ranks of IT innovators swell with opportunistic innovators. However, these firms are less likely to maintain their ability to innovate with IT over time, as systematic innovators do.
So far we have proven that when it comes to IT innovation we can make a distinction between innovative and non-innovative firms. Within the group of innovative firms, we can make the distinction between opportunistic and systematic innovators. Therefore there are three groups: non-, opportunistic-, and systematic-innovators. This observation is consistent with prior theoretical studies arguing that attributes of these firms and their approaches to IT innovation differ.11
Systematic innovators tend to take a long-term view.6,11 In contrast to opportunistic innovators, they tend to have substantial in-house experience in selecting and implementing IT projects and are more likely to have the know-how to evaluate the role of new applications in the context of their organization.11 Systematic IT innovators tend to continue being innovative with IT during periods of rising as well as during periods of falling IT budgets.6 Opportunistic IT innovators, on the other hand, will tend to take a short-term view and are more likely to be driven by the prevailing attitude towards IT.11
Given sufficient time and resources, it is likely that a firm can develop the capability to innovate with IT and to transition from the ranks of non- or opportunistic- to those of systematic-innovators. We were able to find such examples in the list of IT innovators. However, the experience of such companies like Intel2 and Global Crossings7 shows that this is neither instant nor inexpensive. It took Intel five years to transform their internal IT department and reach the level of being considered IT innovative. A similar transformation at Global Crossing took four years. Hence, our prediction: that the development of the capability to innovate with IT is not easily replicated within a relatively short period.
To test this prediction we created four-year rolling windows (such as, 1997-2000, 1998-2001... 2001-2004). Within each one of these windows, we used the following classification: a firm is non-innovative if it did not appear in the list of innovators for any of the four years, systematic if it appeared in more than two consecutive years, and else is opportunistic. Based on these classifications, we calculated transition probabilities from one four-year window to the next. The results, summarized in Panel C of Table 2, showed that the probability is highest that within a short period a firm will maintain its current innovation status. The probability that a non-innovative firm will maintain its status is 94%, that an opportunistic firm will remain that way is 41%, and that a systematic firm will remain that way is 87%. For an opportunistic innovator the probability of becoming a non-innovator is higher (35%) than that of becoming systematic (24%). The probability that a systematic IT innovator becomes an opportunistic innovator is 13%. We repeated our analysis with three and five year rolling windows and the results were similar.
Overall, our results reinforce the anecdotal evidence that we have observed from the cases of Intel and Global Crossing: that transformation from non- or opportunistic- to systematic-innovator is possible but it is not likely to happen in a short time.
This article concludes that the ability to innovate with IT over time is not easily replicated by competing firms. However, there is a limitation that we need to acknowledge. The analysis is based on the premise that a company's listing in InformationWeek is a true measure of IT innovation. While InformationWeek is a well-respected and widely used source of secondary information on IT,10 we cannot confirm that the companies ranked in the list of IT innovators are independently evaluated each year. Nevertheless, since the empirical results are consistent with the anecdotal evidence offered by numerous firms in our dataset, we feel that it does not compromise the overall message and implications for managers.
Based on this study, we advise business managers to avoid investing in the development of the IT innovation capability due to pressure from peers, vendors or other external sources. Systematic IT innovators invest in IT only if it makes sense in the context of the company's business strategy and not because of external influences. If you decide that it makes sense to compete with new IT, you should consider this as a long-term rather than a short-term strategy. Companies in our dataset such as Intel, FedEx, Harrah's, Wal-Mart, and Yellow Road, take a long-term and strategic approach towards IT innovation. By taking a systematic approach, they tend to stay on their IT innovation course regardless of the prevailing managerial perception towards IT innovation and in spite of falling IT budgets. While competing firms may be able to acquire and replicate individual IT innovations that systematic innovators have introduced, it will take them longer to understand and copy the ability to innovate with IT over time.
Our final thought, and our basic piece of advice when it comes to IT innovation, is summarized best by the poet Rudyard Kipling, "... keep your head when all about you... Are losing theirs."
3. Friedman T. Wake up and face the flat Earth Thomas L. Friedman interviewed by N. Chanda, YaleGlobal, 4/18/2005 Accessed on 9/18/2007 from http://yaleglobal.yale.edu/display.article?id=5581
4. Gartner Research. Understanding hype cycles. http://www.gartner.com/pages/story.php.id.8795.s.8.jsp
6. Hoffman. IT innovation interruptus. Computerworld, 9/29/2003; http://www.computerworld.com/industrytopics/retail/story/0,10801,85310,00.html
7. Hoover, N. From bankrupt to innovative in four years. InformationWeek, September 12, 2006; http://www.informationweek.com/story/showArticle.jhtml?articleID=192600624
8. InformationWeek 500 Reports (2002-2004); http://www.informationweek.com/iw500
The authors are grateful for comments received from Bruce Dehning, Michael Ettredge, Younghwa Lee, and Vernon Richardson. We also thank participants, discussants, and anonymous reviewers from the 2007 OCIS Division of the Academy of Management Annual Meeting, 2007 Americas Conference on Information Systems, 2007 Annual Meeting at American Accounting Association, 2007 Canadian Academic Accounting Association, and a research seminar at the University of Waterloo. We would like to acknowledge the financial support provided by the Social Sciences Humanities Research Council of Canada (SSHRC#864-2007-0151-42048) and the University of Waterloo.
©2010 ACM 0001-0782/10/0500 $10.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2010 ACM, Inc.
No entries found