Dr. Urban is applying his forecasting models to the Internet, where the balance of power has shifted to consumers.
Ask consumers to name some of life's top time-wasters and chances are, tasks like shopping, car repair, and home maintenance rank high on the list. From standing in line to waiting for a repairman to arrive, it's the little annoyances that take time away from more pressing concerns like work and family. Although Internet sites promise to cater to busy consumers by offering services such as grocery delivery and bill paying, up to now there's been no such thing as one-stop shopping for Internet surfers.
That may soon change, thanks in part to the work of Glen L. Urban, former dean of the Sloan School of Management and currently professor of marketing management at Massachusetts Institute of Technology. The marketing research pioneer, who has made a career of developing forecasting models to help companies evaluate sales and launch new products, believes that the Internet is going to have far-reaching effects on the development of new products and services, as well as on the way consumers shop.
Urban envisions a virtual world where artificial intelligence "advocates" give consumers information and advice and scour the Internet for the best deals. "Your advocate would be out there fighting for you," Urban explains. It would filter your e-mail for spam and guarding your computer's security. Acting as a trusted advisor, virtual advocates would serve as a concierge, set up medical appointments, or arrange to have your car repaired while you're at work.
If all this sounds more futuristic than the Jetsons, it's nothing new to Urban. Throughout his career, he's been one step ahead of the crowd in market research and product development, and now he's applying his vision to the virtual world. Since the advent of the Internet, Urban points out, the balance of power has shifted from manufacturers to consumers, who are becoming more demanding and more discriminating, empowered by what the Web has to offer. "They have less time, they want exactly what they want when they want it, and they don't want to spend a lot of time looking for it," he explains. In addition, they don't want to have to give up their right to privacy to get it. Third-party agents like Urban's advocates will serve as guardians of consumers' information, as well as handling the time-consuming details of everyday life.
Individual companies, too, will be able to profit from consumers' lack of time and overload of information, according to Urban. Corporate Web sites that develop relationships and offer valuable insight and a sense of community, rather than simply pushing products, will gain consumers' trust - and loyalty - when it comes time for them to open their wallets.
A history of success
This month, Urban joins the ranks of David Ogilvy, Arthur C. Nielsen, George Gallup, and Daniel Yankelovich as the latest recipient of the American Mark eting Association's Charles Coolidge Parlin Marketing Research Award, to be presented at AMA's Marketing Research Conference in San Diego. From his early research in the 1960s using marketing science models to track the impact of information, marketing, and consumer response on sales, to his seminal work in the '70s with friend and colleague Alvin J. Silk, developing a new product forecasting model known as the Assessor, to his latest Web-related research identifying opportunities for new product development, Urban has a long history of making life easier for both companies and consumers.
The packaged goods Assessor model was introduced in 1978. Urban and Silk, currently a professor of business administration at the Harvard Business School, set out to develop a forecasting equation for The Gillette Co. that could reduce the risk and costs associated with new-product failures, using models of consumer behavior and a combination of research methodologies applied before the product was introduced to test markets.
Test consumers were first shown advertising and other communications materials, and then let loose in simulated stores, complete with shelves of new products currently under development by Gillette, as well as existing, on-the-market products, and told to go "shopping." The data collected throughout the buying decision process - how the customers gained awareness of the product, whether they made a decision to shop, to whom they talked about products, and finally, whether they made a purchase - was then statistically manipulated to determine whether the product would sell in an actual store. The Assessor model, which is still being marketed commercially, proved to cut the new-product failure rate by half, Urban says. He later applied it to forecasting the sales of durable goods such as automobiles.
In the mid-1980s, Urban developed Prodegy, a system to identify strategic market opportunities. Although research showed consumers often switched between products within specific segments in a market structure, they would rarely switch between segments. That is, consumers might choose only certain substitutes if their preferred brand is not available. By examining the coverage and duplication of product lines, Urban demonstrated how such consumer data could be used to formulate and test alternative product line strategies. At the same time, in a project that grew out of his work on the Assessor model, Urban analyzed the effects of order on products entering new markets, demonstrating that, if all factors - including price, product preference and promotional spending - are equal, the first movers in a category would garner the largest market share.
Urban's applications have been used by numerous companies, including Noxell, BellSouth, Polaroid, and General Motor's Buick Division. Charged in 1990 with analyzing the sales potential of an electric vehicle from GM, Urban and his colleagues expanded the Assessor model. The new version, called Information Acceleration, used interactive multimedia to create a virtual auto market before the car was built. Potential consumers were given extensive information about the vehicle and placed in a virtual showroom, which included test drives in a laboratory prototype around a shopping mall parking lot.
Urban recalls that although most test subjects said they'd like a quiet, environmentally friendly electric car, their attitudes changed quickly upon learning all the facts through the simulation. Being eco-friendly wasn't enough. Even though the car performed better than people expected - it went 0 to 60 miles in eight seconds - it was only big enough for two and not as reliable as a standard car. "People were afraid [they'd] run out of power," says Urban. And it came with a hefty price tag - $30,000. In order for consumers to sustain interest, he says, the car had to be available at the same price, with the same reliability and the same style as the average car.
Consumer data from various stages in the decision-making process, as well as assumptions about the estimated future price of gasoline and possible tax credits, was statistically manipulated within the Information Acceleration model to determine market feasibility before the production commitment was made. Consumer misgivings and a limited battery technology and recharge infrastructure caused Urban and his colleagues to deem the car a no-go, at least on a large scale. "Our basic recommendation [to GM in 1991] was that they shouldn't [invest billions of dollars] until the battery technology improves," Urban recalls. "And they shouldn't be building a sports car, but more of an econo-van." When the results of the study were later published, Urban forecast sales for 1999 at 1,000 units. In 1998, GM sold only 300.
Even if the news was bad, the model worked with the electric car. But what about the misses? "We were pretty good at picking [products] that paid off," Urban says. But, he admits, some products that his models indicated would die actually turned out to have pretty bright futures. One, a telecommunications device that attached to the telephone and allowed consumers to buy groceries, make dinner reservations, or purchase airline tickets remotely, just wasn't feasible when tested nine years ago. "It was almost impossible for people to get good information [electronically]," says Urban. "It was not so much that we didn't evaluate what was there, but we didn't think the technology would change [so quickly]." Now, with the Internet, everything has changed, except consumers' need for information fast.