Concepts in Motion: Dominant Design

How many different models of mobile phones have you seen? How many of them look alike? Are there one or two models that you think overshadow all others, to such an extent that the rest have to bow down and adapt? If anything, mobile phones seem driven by diversity. While the foundational network has by and large been standardized or even where there is a difference, the network factors in the difference, mobile phones have enjoyed the freedom to pursue idiosyncratic designs.

Look at the automobile business. Even with aero-dynamics reaching plateau (except in very expensive and luxurious cars or sports cars), there is no one model that dominates or to which all else have to conform. The standardization of components has allowed the body to roam free. The one major qualification to this would be engine technology where the internal combustion engine vehicle, a long-standing standard, is under threat from fuel cell vehicle technology.

The striking characteristic of these businesses is that differentiation holds full sway, with none emerging dominant. Not all industries are blessed with or even keen on such flourishing variety. Wherever network effects have a determining influence, something that is commonly accepted (irrespective of how it became so) will prevail as the standard, with others adjusting to this reality.

Falling in line

Hark back to the times when one of the most famous battles in the history of business was between Sony’s Betamax and JVC’s VHS (using some of Sony’s technology). There were two camps. Betamax had Sony Toshiba, Sanyo Electric, AIWA, Pioneer, and NEC while VHS had JVC, Matsushita, Hitachi, Mitsubishi Electric, Sharp and Akai Electric. VHS ultimately became the dominant standard because it was simpler to record, had longer recording times and of course, better marketing. The current simmering friction between Toshiba and Sony over DVD formats has not yet ballooned into such proportions, although Microsoft has signaled its preference for Toshiba while diplomatically maintaining that this doesn’t amount to ratifying Toshiba’s format.

That standards influence our lives can be dated way back in history. In recent years, the emergence of high definition televisions, PC operating systems, modems, Internet browsers and most recently 3G wireless systems are examples of competition to become the dominant design. In 2G, GSM managed to beat CDMA and stay as the dominant design. We notice a difference here, which is significant for understanding the dynamics of acceptance of one design as the dominant design. Unlike the market-dominated US and Canada, the European Community has relied on standards set by the European Telecommunications Standards Institute. The choice of one design as the dominant is a complex choice, in the most descriptive sense of the term.

Dominant design

Welcome to dominant design defined as “a specific path, along an industry’s design hierarchy, which establishes dominance among competing design paths,” according to James Utterback. Since there is competition, the outcome would represent what mathematicians describe as constrained optimization. There is no iron clad necessity that the outcome is strictly controllable, with Utterback and FF Suarez arguing that a dominant design “is the result of a fortunate combination of technological, economic and organizational factors”.

While the concept of dominant design is an interesting area by itself, it is studied for its implications for survival of firms, investment into an industry, number of firms in an industry and therefore the nature of competition, and the kind of efficiency that evolves. As our opening paragraph clarifies, the emergence of dominant design is not a foregone conclusion in all businesses. For example, using data for 63 office products and consumer durables, Raji Srinivasan, Gary L. Lilien, & Arvind Rangaswamy studied the probability and time of emergence of a dominant design and found that dominant designs may never emerge in some product categories.

But where there is a probability of dominant design emerging, you need to grasp which way the wind is blowing. It has a special significance to a broadly defined industry such as IT, where there may be struggle for dominant design status among vertically defined paths. For instance, we already see the struggle happening in service oriented architecture (SOA) between two different ‘standards bodies’. Or in data integration or consolidation.

At one level, the battle for dominant design can be brought down to an unlayered competitive strategy but that would deprive us of a finer understanding of competition in action. There is also the question of whether dominant design is the result or cause of changing competition dynamics. Concepts and practices like dominant design bring a certain sharpness to our understanding of the politics of business. Strategy is not a romantic story, not always, but it perks up our antenna.

Emergence of the dominant

There are seriously divided views on the process leading to dominant design. To JP Murmann and K Frenken, “at the heart of dominant design thinking lies the empirical observation that technology evolves by trail and error and thus entails risks for firms engaged in its development”. This is misconstruing surface images for underlying reality. There may be nothing trial-and-error about the evolution of certain technologies. Utterback sees the process resting on three phases: the product innovation period, the dominant design emergence and the process innovation period. The last phase is important here because once a design has become dominant, there are economies of scale to be exploited which could hinge on process innovation. There are others who accept a four phase configuration, which starts with a discontinuity leading to a ferment throwing up alternative choices with one getting chosen as the dominant design. Quite often, a dominant design is synthesized from more fragmented technological innovations introduced independently in prior products and tested and often modified by users of those prior products. The iPod is a classic illustration of this syndrome.

Marko Hekkert of the Department of Innovation Studies, Utrecht University, Heidelberglaan, and Robert van den Hoed, Delft University of Technology, Department of Sustainability, offer another perspective which broadens our understanding: “Technology dynamics frames competing technologies as a variation and selection process. In variation processes, a multitude of technologies is developed by the technical community as an alternative to the current dominant design. In selection process, through a series of smaller and larger choices made by relevant stakeholders a selection of one technology is made, which becomes the dominant design.” Someone with a very pragmatic bent of mind could well pick and choose these as analytical tools to be used where appropriate.

One conclusion that most agree on is that the dominant design is not necessarily the technologically most advanced. The current trends in wireless technologies together with intense debates over individual companies’ technologies especially Qualcomm is a telling illustration of this realistic portrayal. When Sprint-Nextel chose WiMax over Qualcomm in what has been billed the biggest support for WiMax, Sprint-Nextel said that while they admired Qualcomm’s technology, they chose WiMax because it would help them to cut the time to market. It need not all be politics. When it is, it is loud and clear.

Take 3G, where the world is eagerly awaiting what is going to be China’s choice. As potentially the largest market, the Chinese decision would have serious implications. The choice is very complicated given Chinese conditions, with GSM, PHS and CDMA all in use, and with four major telecom operators. The Chinese Ministry of Information Industry favours TD-SCDMA while operators are keen on WCDMA. Characterising TD-SCDMA as the daughter and the operators as prospective sons-in-law, one writer says that the marriage of TD-SCDMA with operators will be an arranged marriage. No love permitted here. Standardization here would mean a framework that enables production of goods that are interchangeable or compatible. As Ericsson put it, “The question of 3G deployment is not a technical issue any more, but a regulatory and economic one”. (Quoted in ‘From Pipeline to Stovepipe – unbundling in the mobile communications industry’ by Pablo Valiente of the Stockholm School of Economics).

Process innovation: The politics of perception

As we mentioned above, once a design has become dominant, innovation usually turns to process. One of the classic examples of this is Dell Computers which, Carly Fiorina, then HP’s CEO, dubbed the most innovative logistics company. She wasn’t being charitable though, but Dell certainly didn’t mind the description. The PC design was already dominant and Michael Dell didn’t want to overturn that, overturning instead the way the business was run.

There are influential arguments now that the $2.1 trillion IT industry is in the process innovation mode. Is such a view innocuous or is coloured by unarticulated undercurrents, which we must excavate to unearth the politics behind? IBM has already proclaimed its belief in process innovation and has gone on record to say it doesn’t see any major breakthroughs in products. In this specific instance, process innovation has come to mean standardization, a view supported by the current digitization technology that is at the heart of services transformation: processes are now formalisable, codifiable and therefore repeatable. In a subtle fashion, IBM is creating the path to a future dominant design, understood here as the ‘one-size-fits-all’ syndrome.

We must maintain vigil, but it has to be tailored to specific segments.

Didn’t make it

The dominant design can be a simple data format or a DVD. We all know the current friction between two DVD formats – one from Toshiba and the other from Sony. In the case of health care, it could be the format for capturing health records, such as the electronic medical records (EMR), an idea that has been around since the 60s. However, this was not just an electronic record; it called for a fundamentally different approach. Larry Weed introduced the notion of Patient Oriented Medical Record into practice at a time when doctors were content to record their diagnoses and the treatment they provided. Weed’s innovation was enabling a third party to independently verify the diagnosis. In 1972, the Regenstrief Institute developed the first such record. While widely hailed as a major advance, it also faced stiff resistance from doctors. Now more than three decades later, EMR has still not taken off, its use confined to some 20% of hospitals in the US (and less in Canada) and 5% in clinics. (Data as of 2002)

Network, network, network

We are part of an interconnected whole but a network which goes by its logic and within which products must find their place. As Robin Cowan puts it, “Network effects are thus at the heart of many new technologies”. Any firm keen to unseat a current dominant design has to address the question of switching costs, often a serious impediment to embracing the new.

Collateral or specialized assets to play a crucial role but have a two-way relation. It makes more sense for a firm to acquire collateral assets once a dominant design has set in, but the opposite relation may also hold: a dominant design will tend to stimulate the creation or acquisition of collateral assets, which in turn will strengthen its dominance. It is instructive to reflect on the battle between SAP and Oracle in the ERP space especially on their various acquisitions to fill gaps in repertoire from this perspective.