Retailing and the InfoBahn: Predictions and Advisories

Marc Demarest

Marcia V. Wilkof
School of Business Administration
Philadelphia College of Textiles & Science
Schoolhouse Lane and Henry Ave.
Philadelphia, PA 19144

May 1995



Technology is dramatically and fundamentally changing industries and organizations. This paper is about the changes looming just on the horizon for the retail industry, in particular the clothing industry. It describes two visions, one of mass customization of clothing, the other of electronic shopping via video malls. Both technologies are described and some of the implications for the industry, relevant organizations and the consumers are explored.



In a recent address to the American Magazine Conference, Barry Diller, the chairman of QVC, described his decision to turn QVC -- the world's best-known electronic retailer -- toward software development:

It's only with patience that you can develop fluency in a new medium. Let me tell you what I mean. I have invested research and development dollars at QVC in a technology called Smart Agenting. Smart Agenting uses a computer to construct a profile of you -- your tasts, habits, your likes and dislikes. You build that profile by answering questions, endless questions with precise grades of sensitivity. Then, with that profile stored away in its microprocessor, that box does what none of us has the time and energy to do: it searches every corner of the information universe. It searches wide and targets narrow. It comes back to you in a split second. And it offers the goods and services, and information you want.

While it is common for us to think of retailing as the deployer of software technology developed by other organizations (and perhaps as the developer of software for its own internal operations), it is quite uncommon to think of a retailer as a developer of software for the broad market. Diller's decision to support a significant software development effort in a retail organization offers us a compelling and intriguing glimpse at a concept that is starting to gain currency in organizational design circles: the notion of a design center firm that, in the context of a large network of suppliers and manufactueres organized into a value chain, uses its unique and robuest understanding of a particular customer group and a particular delivery medium to control and direct the operation of that value chain. In the case of QVC, Diller's strategic intent is clear; to parlay QVC's mastery of the analog television medium as a channel to market into control over the next electronic channel to market for soft goods: the information superhighway, or, as well shall call it here, the InfoBahn. Diller's control over that medium, according to this strategy, lies in his control over the software that purchasers use to locate goods and services, particularly in the algorithms and psychodemographic assumptions that underpin the questions that the smart agent technology asks the purchaser in order to form an accurate picture of the purchaser's needs and desires.

This paper focuses on the area to which Diller's decision calls our attention: impact of technological changes that are looming on the horizon in the retail industry, primarily but not exclusively in clothing. In this essay, we present visions of what shopping may be in the 21st century, and raise squestions about the nature of the value chains that will serve the electronically-enabled retail customer to make these visions realities. We begin by describing the technology in use, and then explore its impact on the retail industry and the workplace.

Images Of The Future

Digital Mannequins

Enter a Scantron store. Swipe your credit card, and enter a booth where, using both lasers and soundwave technology, a computer scans every part of your body, creating an exact image of each component, imperfections and all. Outside the booth, collect a small form-factor card with digital models of your body image stored in it, as well as the software required to display, from any angle, this data in aggregated form -- as an arm, a head, or a body -- on your personal computer.

Mass Customized Clothing

Go to your favorite retail clothing store to pick up a new suit. Using the store's computer system, select the suit style and fabric that suits you. Test an electronic mock-up of the suit, built on the fly by the computer, against a variety of accessories: shirts, ties, belts, shoes. Enter some information about your preferred fit: slightly big in the shoulders, tight waist, somewhat long in the legs. Insert your body image card in the computer system, and view yourself in the suit. Click a single button, fill in delivery instructions, and transmit the entire data set -- suit style, fabric choice, your digital mannequin, and your epersonal digital signature and electronic bank account information -- to a manufacturer's factory, where your job is queued for computer-integrated manufacturing (CIM) systems that, within 12 hours, have produced your suit to specification. Within 24 hours, via overnight shipping, the suit has arrived where you asked for it: your home, your office, or the next stop on your two-week business trip.


Turn on your TV and flip to the "video mall" channel with the remote control. Using a wireless keyboard and mouse, drive across the InfoBahn to your favorite electronic mall. A tops-down scehamtic of the mall, showing the electronic store layouts, appears on your TV screen. Click a button, and a list of the stores, sorted by type, appears in place of the schematic. Click another button, and you start a slow-paced "walk through" of the mall, glancing in the stores along the way. When you come to a store of interest, you "turn" into the store and browse around until you see an item you're interested in. You then click on the item, the TV screen fills with a rotating, three-dimension image of the product. You decide to purchase it by clicking the button at the bottom of the TV screen maked "BUY." The TV's set-top box transmits your digital signature and information about the store to your bank, which in turn uses electronic funds transfer (EFT) technology to cut a purchase order and send it to the retailer, with advisory instructions designating your home address as the ship-to point.

The Future? Or The Present?

All the technologies used in the scenarios described above are commercially-available today.

Digital Mannequins

Laser technology for body mannequins costs between $250,000 and $450,000 for a laser scanner, depending on how much of the body is to be scanned. Several organizations have been working on this technology. According to Joseph Ott, Managing Director of the Textile Clothing Technology Corporation (TC-Squared), their version is an eight foot square booth that has lights and cameras in each corner pointed toward the center (Charles, 1994). People would put on a white tight-fitting body suit so that computers can calculate their exact shape based on how the light falls against their bodies. Clothing is then custom-made to fit.

PCMCIA technology for personal computers allows between 2 and 10 MB of digital information to be stored in a device with a form factor of a credit card, and provides a standard interface for connecting such "credit card" devices to any computer system.

Levi-Strauss announced a program in the fourth quarter of last year that might be a benchmark in mass customization and customer focus, and a harbinger of the future of retailing. In a nutshell, Levi-Strauss announced that its customers, by the end of 1995, would be able to walk into certain retail outlets, have their lower bodies scanned by laser, and, from the measurements captured by the scanning equipment, have a custom-cut pair of Levi's made for them. Furthermore, Levi-Strauss announced that whenever a customer who had been scanned needed a new pair of jeans, that customer would be able to call a toll-free number and request a new pair of pants. The catch, and it was a subtle one, escaping almost all the commentators who covered the announcement, is this: the customer must always go to the retail outlet where the initial scanning was done to pick up the custom-cut jeans.

Electronic Mass Customization

The integration of computer-integrated manufacturing environments with the Internet is already in the late existence-proof stage. It is already possible for example, to get custom-built integrated circuits or machined metal parts developed and delivered in a matter of days by sending special instructions, in the form of a mail message, to Internet-connected CIM environments at various US universities. The specified components are manufactured and shipped automatically.


There are today over 2000 digital malls and retail outlets using World Wide Web technology on the Internet alone, selling everything from cars (, car parts ( and real estate (a href=" to men's suits (, perfume ( books (, music ( and marital aids (

Although placing an order over the Internet today is a risky proposition, involving the exchange of credit card information in unprotected ways (Lang, 1994), there are four standards for secure digital currency proposed and two of the four are currently operational on the Internet. Microsoft and Visa International will offer a fifth when Marvel debuts with Windows95 (Murray, 1994). Mosaic Communications, developers of the most popular commercial interface to the World Wide Web, has entered into an agreement with MasterCard to implement the payment protocol described in the video mall example above, with likely commercial availability in early 1996. It is expected that a de facto standard for digital currency will be established by the end of 1996.

Reliable, secure digital signatures are already possible For example, Pretty Good Privacy (PGP), is available free on the Internet, and has about 250,000 users worldwide today.

Andersen Consulting's retail division has opened The Retail Place, a demonstration facility targetes at its retail clients that exhibits alternative technologies of the future (Holton, 1994). The home shopping station, set up as a living room, gives people the opportunity to interact with a "video mall." A remote control moves the mall; the consumer can "walk" down the corridor clicking on stores they are interested in. They then get fully animated descriptions of the store's products. Andersen believes that customers would identify a product they want to purchase, enter their credit card number and indicate where they will pick up the product or where it should be shipped.

Nordstrom offers its customers today a 24-hour electronic-mail (e-mail) shopping service. A collaboration between Nordstrom's Seattle, Washington branch, MCI Communications Corp., ConnectSoft, the service, dubbed Nordstrom Personal Touch America, or NPTA, provides "personal shopper" assistants and a two-day Federal Express delivery of the selected goods. The new service accepts Nordstrom's in-house credit card, Visa, Mastercard, and American Express.

The Context

There are of course, all kinds of technical and human barriers that prevent the rush to the world we have outlined above. For example, Time Warner's Full Service Network trials in Orlando, Florida have cost $750 million just to wire 4,000 homes, and estimates are that equipment will cost several thousand per household and wiring several thousands more (Shiver, 1994). Similarly, noteable early network-based e-malls like MecklerWeb failed utterly to stimulate the level of digital revenue they expected (The Wall Street Journal Oct 20, 1994 pB7(W) pB7(E) col 3).

Given the expense associated with readying television as a platform for interactive shopping, given a decided mismatch between the socioeconomic groups with high levbels of disposable income, and the socioeconomic groups for whom television shopping is appealing today, and given finally that the InfoBahn is up and running today, and used by the sociodemographic groups with the highest levels of disposal imcome, we believe the networked personal computer is a more likely platform for interactive shopping.

The Infobahn As Electronic Marketplace

Ultimately, the InfoBahn is nothing more than an electronic channel to market, although it exhibits very different dynamics from the channels we are used to in conventional retailing. To get an idea of how large the consumer electronic purchasing market will be by the end of the century, consider these trends:

  • Today, some 40 million people worldwide (Murray, 1994) and 10-12 million U.S. homes are connected to the loosely-coupled international set of networks centered around the Internet. By the end of the century, at present rates of yearly growth, 85 million U.S. homes will be connected to that same network of networks, which, by that time, will have gotten tighter, faster and easier to use.

  • Today, the average connected household makes about $85 in electronic purchases, half of which is accounted for by the cost of the access service (America Online, Prodigy, CompuServe, Internet point of presence providers). Even if we assumed that the average household's electronic purchases only reaches $500 per household by the end of the century, $490 of that will be purchased goods and services, given the current rate of decrease in the cost of access (driven by Microsoft, among others, which threatens to cut the cost of InfoBahn access to virtually zero as part of its Marvel service program). Eighty five million homes, spending only $500/year is $40 billion dollars a year. And our calculations are conservative. Management Horizons predicts that the market for electronic retailing by the year 2004 will be $150 billion (Shiver, 1994).

  • The World Wide Web, the most successful mechanism developed to date for networked advertising, sales and fulfillment, has grown to well over 1,000,000 Internet nodes, some 20% of which are commercial, after less than three years in operation.

  • IDC and other information technology market analyst organizations are forecasting the total market for consumer-initiated electronic sales at anywhere from $10 to $21 trillion dollars by the end of the century.

While the absolute numbers are not relevant and may be inaccurate at this juncture, and while telephone and cable companies' trials have been largely inconclusive to date, there is a whole set of younger, smarter psychodemographic groups going online. These groups have both the inclination and the wherewithal, in terms of disposable income, to purchase the majority of their goods and services electronically. And

children, who are the most computer literate of us all, will view this use of technology as an integral part of the landscape of their lifestyles, as they grow up interacting with Walt Disney and Random House multimedia software products the way a previous generation used books and board games.

In much the same way that many people do the vast majority of their banking through ATMs, finding it foolishly unthinkable to waste their time in a bank, today's children may very well have the same attitude about e-shopping -- why would anyone waste their time walking from store to store, getting undressed and redressed to try on outfits, when its easier, more effective and more efficient to shop on the InfoBahn?

The Implications

We believe that the primary implications of the emerging electronic channel to market are those that affect the arrangement and power relationships in the retailing value chains that serve customers today, particularly the contenders for the "design center" of that value chain: the firm whose understanding of the market served places it in the position to control and orchestrate the activities of the other firms participating in the value chain.

The current value chain structure of the retail industry, in general, is that textile manufacturers design and produce fabric. These companies sell their fabric to apparel manufacturers who design and produce finished garments which, in turn, are sold to retail outlets. The impacts on this industry as a result of the new technology will challenge our conventional ways of thinking about and organizing the production and purchase of clothing, suggesting a range of "strategic advisories" to companies in this industry.

The best study of this design center model of retail value chains is Clark & Stanley's study of Marks and Spensers (Clark & Stanley, 1989). In that study, Clark and Stanley described an extensive value chain, driven by Marks and Spensers and including raw materials manufacturers, clothing manufacturers (both private label and brand), and the UK central government, all orchestrated by Marks and Spensers, whose intimate understanding of the particular upmarket it served gave it control -- full or partial -- over the activities of other value chain participants, control that was so complete as to threaten the economic viability of some of those participants. Commenting on the study, Clark and Staunton concluded that:

[The] example illustrates the case of existing chains which are the outcomes of past investment and foresight, but does not highlight how economic restructuring is altering those and providing new opportunities. Clearly existing chains are also political and power systems in which vested interests occupy certain positions. It must therefore be anticipated that new chains are emerging, especially across national boundaries.

Other much-used examples of retail value chains where the design center is held by the retailer include Wal-Mart and its much-discussed use of category management processes driven by its vendors, and the uses made by Benneton of GEISCO networking services to link its operations to those of its suppliers. But it is by no means the case that, in all retail value chains, the retail outlet chain unit is the design center; consider, as a case of the designer-as-design-center, Liz Calibourne's control over both the manufacturing elements in some of its value chain as well as the raw materials vendors who supply those manufacturers.

The fundamental promise of the InfoBahn -- a solution to the logistics costs problem in the traditional multi-tiered retail channel -- is undeniable. If, on average, recovery of the cost of moving made goods to the point of sale accounts for 50-75% of the retail price of an item, then the rationale for forward integration through the InfoBahn is compelling to every player in the value chain, from the design house and the manufacturer through to the wholesaler and distributor. The only factor mitigating against massive forward integration is the fact that historically the retailing agent firm has functioned as the designer center firm in the value chain: managing the analysis and definition of offerings based on its (relatively speaking) intimate understanding of customer needs.

So, what the availability of the InfoBahn and associated technologies does, generically, to the retail industry is open up, in a broad way, the battle for control of the design center of retail value chains. It allows both manufacturers and designers to forward-integrate, past their wholesale and retail channel elements, and touch the end-consumer directly in intimate ways. It challenges the retail outlets, traditionally in the bext position to command the design center role in the firm, to hold on to their control over the needs and tastes of psychodemographic groups. And, for all participants, it changes radically the cost structure of the retailing business.

It is too early to develop any comprehensive model of this change; we have no long-lived examples of successful eletronic retailing and only anecdotal evidence of firm intentions and programs. Nevertheless, we would like to examine some of the issues raised by this looming "battle for the design center" focusing on the technology, the industry, the organization and the consumer.

The Technology

There are no significant issues around the availability of the technology required for retailing on the InfoBahn: all critical technological components are in place and operational. Instead, the technological issues are those of standardization and penetration.

  • Standardization. Given that there are competing technologies for digital mannequins, digital signatures, e-cash and the protocols that link electronic retailers, banks and consumers to one another, there will be some significant lag between the electronic retailing infrastructure we have in place today -- multiple conflicting, non-interoperable standards -- and the availability of a stable infrastructure based on commonly-accessible standards. Firms who embrace electronic retailing during this "stabilization" period run the risk of implementing their retailing technology on top of other technologies that are weeded out in the standardization process. We would at this time predict that the technologies proposed by Mosaic Communications, First Virtual Corporation, MasterCard, Visa and Microsoft are likely winners in the standardization wars. Similar problems exist in the technical integration of manufacturing equipment into this environment: no standards exist today for the description of apparel items to CIM equipment, and we are aware of no efforts to produce such standards.

  • Penetration. In the mid-term, it is likely that one or a few psychodemographic purchasing groups will be closely linked to the InfoBahn, and that products and services not relevant to those groups will find no significant advantage in electronic retailing. For example, while Wired, Time, and the various computer industry trade magazines are broadly read on the Internet today, we find the idea of an online Reader's Digest lucicrous; there is simple no overlap between RD's readership and the psychodemographics of the Internet today. Similarly, there is a viscerality to retailing today: consumers like to touch, and try on, what they purchase. We expect a rapid migration of mail order clothing companies to electronic retailing, since mail order is a successful channel with no viscerality, but we also expect that a generational transition will be required before the majority of consumers do not need visceral experiences of apparel and other retail items prior to purchase.

Industry Relationships

  • The relationship between textile and apparel manufacturers. Does the relationship between textile and apparel manufacturers change with this new technology? Probably not. The former are good at designing and producing fabric, are capital intensive and have more of an engineering culture. The latter are good at designing and manufacturing finished garments, are labor intensive and have more of a production culture.

  • Apparel manufacturers. How will manufacturers handle mass customization which requires a shift from mass production to unit production? This shift would most likely result in radically-increased use of CIM equipment and production systems, with concomitant in the human resource skill mix required by the firm.

  • The relationship between apparel manufacturers and retail stores. Department stores have moved to develop their own private store labels, and apparel manufacturers have moved to open their own retail stores, both in an attempt to cut out the other and thereby increase their own profit margins. Both could view mass customized, eelctronically-order clothing as another marketing channel, similar to the way they view home shopping. Apparel manufacturers who actually do their own manufacturing -- as opposed to contracting it out -- may be in a betterstrategic position to forward-integrate past their current retailing partners because they control manufacturing. We would see Levi-Strauss' emphasis, in its scan-technology announcement, on having its customers pick up their custom-made clothing at retail stores as an attempt to convince its retail partners that it has no strategic intent to forward-integrate. For those companies who contract out their manufacturing, the risk exists of back-integration, as retail outlets intent on retaining control over the design center purchase manufacturing capability to achieve control over the means of production.

  • The relationship between apparel manufacturers and video malls. Video malls could be seen as simply another channel for the presentation and sale of merchandise, like retail stores, discount outlets, catalogs and television shopping networks. Like the latter two, video malls provide more challenge for quick response and essentially no requirements for keeping inventory on hand. There is no sensible case to be made for providers of video or network floor space to backward-integrate into manufacturing; hence we predict no conflict between electronic floor space providers and the manufacturers and retailers who use this infrastructure.

The relationship between retail outlets and video malls. Retail outlets may want to form strategic alliances with video malls in the same way that some now have a catalog business or utilize home shopping channels. Early e-shopping malls based on World Wide Web technology have yet to demonstrate that such partnering synergy exists, however -- they function much more as electronic billboards than as actual sales outlets.

  • The catalog business. Mail order catalog sales is the most attenuated kind of retailing we know. It is also the most efficient, and if recent studies are to be believed, increasingly preferred by households with high disposable incomes. Presumably, if Nordstrom's could offer its hallmark customer service and operate as a mail order house, it would. (This brings up the issue of proximity. People generally have the notion that we have to be face to face, viscerally in touch, to have high touch service. This is true only for our generation, not our children.) Consider a mail order catalog operation with $100 million in annual sales and $85 million in annual costs, leaving a not atypical (some would say quite healthy) $15 million in margin. This mail order house is cutting-edge; it maintains little or no inventory, ordering from manufacturers as it receives orders, perhaps not even repackaging and reshipping. When we look closely at this firm's costs, we discover that well over $50 million of those are associated with demand creation, mostly with the production and distribution of their catalog, which is issued in 500,000 unit runs four time a year and has a buy-rate of about 4%. The cost to produce that same catalog today, using the publication technologies associated with the World Wide Web, is about $500,000 (including production and Internet access), approximately 1% of the print-based costs. In addition, electronic publication of the catalog:

  • allows firms to publish perpetually, changing the catalog contents on a daily or hourly basis.

  • allows firms to publish fully multimedia catalogs, including walk-throughs of rooms furnished with products, people modeling clothing and children playing with toys.

  • allows the firm to tailor the product to individual readers as they "log on" to the catalog based on past purchase history, psychodemographic data, etc..

  • allows the firm to watch readers interact with the catalog in real time and gather more information on reader preferences as well as broad product line appeal.

Organizational Concerns

  • Delivery. There is a central distinction in our basic model of electronic retailing between demand creation, the order that demand creation ideally produces and the fulfillment of that order with actual products or services. Although it is blindingly obvious, it is worth restating: Benneton cannot deliver an angora sweater over the phone line any more than General Electric can deliver a washing machine via a coaxial cable, no matter what hallucinations Bill Gates may have on those dark winter nights in Redmond. Until recently, those laws of physics kept manufacturers of soft and hard goods off the InfoBahn, but there has been a shift in the last year as more soft and hard goods manufacturers recognize that they can take orders electronically from end-consumers even if they have to fill these orders "over-the-road" instead of "over-the wire." Will retail stores require that customers pick up their customized clothing at a retail outlet or can they simply have it sent to the home or office? There are arguments for both and perhaps the customer will be given the choice. Having customers pick up the clothing at retail outlets assures that a qualified professional is present who can assuage concerns about slightly ill-fitting clothing, immediately arranging to make any needed alterations. This approach prevents customers from being upset about clothing received through the mail that does not fit exactly as imagined. The concern is the damage that could occur to customer goodwill, satisfaction and loyalty.

  • Will the Eddie Bauers and the Land's Ends be more willing to ship custom made clothes directly to the customer because they have years of experience with catalog sales?

  • Will other apparel manufacturers view this technology as a way to by-pass retail stores altogether and is that desirable?

  • Return policy. If a customer does not like the clothing, for whatever reason, can it be returned? It is most likely that retailers would have return racks of "as is" clothing grouped by general sizes that would be discounted. While some alterations may be required, current thinking is that these returns could fairly easily be sold.

Consumer Issues

  • Ease. Many consumers will welcome the variety of technologies described in this paper. Studies have shown that many people would welcome body scanning if it leads to purchasing clothes that are better fitting (Knight & Cassill, 1994). And electronic shopping on the internet or home shopping channels has continued to increase in popularity.

  • The feel of it all. While many customers welcome the ease of electronic shopping some still voice preferences for being able to feel the fabric and experience what it looks like before making a purchase.

  • Leisure time. One of America's favorite pasttimes, however, is shopping which suggests that it is unlikely that electronic shopping will replace traditional forms of shopping. People worried that when the VCR became a common household gadget,no one would go to the movies ever again. We have seen that that is not true, but we have also witnessed the demise of grand old movie theaters and the rise of the multiplex theaters. In retail we have paradoxes of specialty retailing and outlet stores; small individualized boutiques and huge discount changes; regional chains and moves toward mega-mergers. In lots of locations retail has moved from the center of the city to suburban malls and technology now and increasingly in the future allows retail to move into the home. These contradictions and paradoxes are strategic challenges in retailings future.


What we have tried to do, in this paper, is to spark discussion about the convergence of several technologies that we believe will change the way we shop in the not too distant future. To summarize the major points:

  1. Technology is on the horizon that will make mass customization more common and more desirable than mass production.

  2. The InfoBahn will allow manufacturers with brand equity and design center aspirations to forward-integrate through the retail channels they use today and sell directly, inside a mass customization paradigm, to the end customer, offering lower prices and getting higher margins on a per unit basis: a fairly compelling proposition.

  3. Manufacturers of soft goods and some retail stores will enter into competition with mail order catalog operations that sell their products today.

  4. Retail stores will have to go online in very sophisticated ways to retain their traditional role as the design center firm in the value chain, and will -- if they are successful in holding on to that key position -- be able to offer benefits similar to those the manufacturers see, because of reductions in fixed costs associated with the transition from physical to electronic point-of-sale models.

  5. Retail agents that do not hold on to the value chain design center become, for those involved in electronic shopping, "dealer showrooms" for consumers who need visceral experience of the product before purchase. Over time it is expected there will be fewer and fewer of these.

It is hard not to give into hyperbole, to proclaim ourselves on the cusp of a revolution as disjunction and shattering as that of the Industrial Revolution. One thing is certain -- like the Industrial Revolution before it, the InfoBahn Revolution will not announce itself. We should expect no watershed, no defining moment, no formal declaration that the age of electronic retailing has arrived. As with many of the technology-based markets, one insignificant event -- the last, anonymous, silent grain of sand in the balance -- will likely be enough to tip the scales quickly (and painfully) in favor of electronic channels to market. Some people would argue that that grain has already hit the balance -- in the past month the two largest players in the high technology market, Oracle and Microsoft, made significant announcements with respect to the InfoBahn. Microsoft declared, in typical fashion, that it intends to own the InfoBahn in toto, and Oracle Corporation -- the only firm likely to thwart Microsoft's world-dominance strategy -- announced that it will sell its software -- retailing from $50,000 to $350,000 -- electronically via the World Wide Web.


Last updated on 06-22-97 by Marc Demarest (

The authoritative source of this document is