The Winner's Circle

BOSTON (05/31/2000) - Clearly, one of the diseases of the Internet Age is hubris. Symptoms involve "Cashing out quick before the investors understand what's going on here."

A subtler and more interesting symptom of internet hubris is the "we can have our cake and eat it too" business model. With it, the thinking goes, the internet is such a powerful tool that companies can ignore time-honored business choices and have it all. Low cost or good service? Consumer markets or business-to-business? Companies have always had to make such distinctions based on good business sense, and strategy textbooks have rationalized such decisions since Michael Porter was in diapers. Sure, a company can have different business models under one corporate roof, but conventional wisdom dictates that they be segregated across different business units.

Now a lot of organizations are feeling constrained by such thinking; they want to simultaneously adopt both alternatives within one business unit. Strategic focus no longer matters in many internet business models; all that matters is being on the internet and having customers. Maybe I'm just a hopeless relic of the old economy, but I still feel that on many issues of internet strategy you've got to declare which side you are on.

In the business-to-business world, many e-commerce hubs, exchanges or networks are trying to be all things to all customers. Are they places to buy or sell surplus supply on the cheap, or are they going to create tight, collaborative relationships between buyers and sellers? Are they for hard-core buying or casual chatting? My sense is that if you try to do all of these things, you won't do any of them well. The cost structures, processes and management styles necessary to do one type of business well will leak into areas in which they don't fit. Despite the miraculous qualities of the internet, straddling business models within a business unit will ultimately lead to painful splits.

The recently announced mega-exchange for the Big Three automakers has elements--at least in the aspects of its design that have been announced--of both cheap source for marginal purchases and integrated core supplier market.

The companies expect tens of billions in cost savings, which suggests that suppliers will be pitted against each other to submit lowball offers. But reports also suggest that some of the exchange's applications will include advanced planning and scheduling, demand forecasting and demand collaboration.

That suggests close integration with suppliers. Has it occurred to anyone in Detroit or Stuttgart that these purposes might not be easily reconciled?

It has occurred to Toyota, whose participation in the exchange was actively sought by the other three partners. But in a Financial Times article on the exchange, Tadaaki Jagawa, Toyota's head of procurement, noted that, "Our parts are not purchased through a bidding process. We buy them by building a relationship with our suppliers over time." Analysts suggest that Toyota's close relationships with suppliers are critical both to its high quality levels and its shorter new car development cycle than those of U.S. or European automakers. Toyota will participate in the exchange, but only on the margin: buying nuts and bolts and office supplies through it. In other words, Toyota views the exchange as a place to save a few bucks--er, yen--on stuff that really doesn't matter much to its business. Supplier integration with its business processes will take place not over the internet but through extended face-to-face negotiations and linkages between proprietary systems.

One reason that e-networks need to choose between low price and integration is that integration is surely going to be expensive. It took a decade and hundreds of meetings, for example, for the automobile industry to develop EDI standards and integrate with its suppliers using that technology. EDI is sometimes described as an expensive technology, but its cost pales in comparison to the human costs of agreeing on information and process standards. Suppliers can justifiably argue that they are supposed to post rock-bottom prices to get business on an exchange, while having to spend megabucks on integrating with car companies. The automakers will have to work closely with suppliers if they're going to meet their goals of rapid new car development and build-to-order manufacturing processes, so they won't be able to buy from the lowest-price supplier on the exchange. I'm predicting that the U.S. and German companies involved in the exchange will end up buying only commodity stuff, just as Toyota plans to do.

The Vertizontal Network

Another "have your cake and eat it too" situation can be found in the race to make vertical networks horizontal, and at the same time to augment e-content and e-community with e-commerce. Nobody's satisfied with their little niches.

The trend today is to expand single vertical markets into multiple ones.

VerticalNet is the paradigm of this strategy, having expanded into more than 50 different industries. That worked fine when all it offered was trade publication content and advertising, which was its original business model. But then it became clear that all those industries would eventually be offered e-commerce transactions through exchanges, and VerticalNet needed to grow.

Transaction-based revenues are often more lucrative than relying on advertising. However, it's going to be tough for VerticalNet to convince content viewers to become exchange participants. Readers of, for example, have often built up relationships with industry colleagues over years at trade shows, golf games and other face-to-face settings. Moving the relationships online may simply not work, or may take too long for VerticalNet to profit from them.

Other companies that previously specialized in one kind of product or industry are expanding. Chemdex, which sells laboratory equipment online to the life sciences industry, has announced that it will diversify into other types of products and markets. In fact, the company is no longer Chemdex--that's just one of its offerings--but Ventro, a general provider of B-to-B marketplaces.

Why make the shift? Because there are only so many dollars one can wring out of distributing beakers and Bunsen burners.

But what do Chemdex's customers lose when Ventro goes multivertical? Perhaps nothing initially. But eventually the people who specialized in the laboratory and life sciences markets will be asked to focus on other kinds of markets as well. The knowledge of industry people, processes and products will get watered down. Wall Street analysts may applaud such diversification, but I expect that over the long run customers will decry it.

In short, this is yet another area in which the rules of e-commerce will converge with the rules of commerce: Stick to your knitting. Don't get stuck in the middle. Strategy is about choices. Isn't it great that we can get new life out of these old business clichs by applying them to electronic commerce?

Tom Davenport pontificates at both Andersen Consulting Inc.'s Institute for Strategic Change and Babson College. He welcomes reader comments at

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

More about AndersenAndersenAndersen ConsultingChemdexToyota Motor Corp AustVentroVerticalnetVertical NetworksWall Street

Show Comments