E-marketplaces: what went right and wrong

But that's just what happened earlier this year. Public e-marketplaces, which only 18 months ago were going to change the way we thought about e-commerce, are limping into semiretirement as private marketplaces move in to take their places and as the companies that originally preached their value reverse the original prognostications.

"As far as public marketplaces go, the ones that deal in highly commoditized products survived," says Bob Shecterle, vice president of e-commerce strategy for PeopleSoft. "When you move to goods that are outside those industries, it becomes very difficult for the public marketplaces to provide value."

E-marketplaces were designed to connect businesses by matching buyers and sellers regardless of existing relationships or size. Companies that wanted to set these up bought software from Ariba, Commerce One, CommerceQuest, Computer Associates International, Metiom, and PeopleSoft, among others, and offered a venue where businesses could trade, auction, or sell merchandise and services.

In a perfect e-marketplace, companies could buy and sell goods and services regardless of a business's size, location, or existing relationships. For example, if a company were selling transistors, it could post them on an e-marketplace and find out the demand and the market price. Businesses as large as IBM and as small as the local mom-and-pop computer store could purchase the transistors--no sales call required. Since e-marketplaces would theoretically connect to supply-chain management software, a manufacturer's production plant would be alerted to make more of a specific item as soon as a sale went through. Production could be scaled back for those items that didn't sell. Sellers would be able to use e-marketplaces to unload excess inventory, while buyers could go out and purchase what they needed on the spot.

Though a nice idea, that's not what happens. Sure, several vertical e-marketplaces--those that cater to narrow industry sectors rather than an entire industry, and most notably those used to sell chemicals, metal, and other raw materials--emerged with characteristics of the original e-marketplace vision. But numerous others flamed out after burning through literally hundreds of millions of dollars.

Take Transora, a marketplace for packaged-goods companies. Transora's investors include Bristol-Myers Squibb, Campbell Soup Co., Kraft Foods, and Unilever. To date, the e-marketplace has raised $240 million by asking for $500,000 to $15 million from 54 of its represented companies. All of these companies use the marketplace to buy and sell services and goods ranging from honey to truck space and packaging. With their investments, they have a financial incentive to be sure that the e-marketplace is used often. Transora has used part of the money it's raised to make the site more valuable by adding community, content, marketing tools, and supply chain services.

Other e-marketplaces, such as the World Chemical Exchange, or ChemConnect as it's commonly known, are successful, because they provide a way to buy and sell products that have few variations. Companies can go after as much or as little of a product as they need and get it from whomever has it readily on hand. Since commodity pricing is set by supply and demand, companies aren't locked into a set price at all times. "These kinds of products have a rating system, so people always know what they're getting," says Leo Lipis, senior analyst for research firm IDC.

Although this might work well for iron ore and corn, the open marketplaces, with many different companies participating, aren't necessarily well suited for purchasing items that are essential to operations, such as LCDs for Dell computers. Dell needs guaranteed delivery by a certain time, and that's not always possible in an e-marketplace.

That type of product is better suited for a private online marketplace, which is run by a particular company for its customers and suppliers. In a private marketplace, no bidding or negotiation is necessary. The service's main goal is electronically connecting companies that regularly do business with each other.

The transition from public to private e-marketplaces is picking up steam, because private marketplaces make it much easier for buyers and sellers to work together without sacrificing quality or profits. RestaurantTrade is a prime example. The New York-based private marketplace caters to restaurants, bringing chefs together with hundreds of wholesale food and equipment suppliers. One of the most useful features about the marketplace is that restaurants can use it to order from suppliers that aren't officially part of RestaurantTrade. Restaurants can specify what they want online, and the orders are then faxed to the suppliers. Mina Newman, executive chef at New York steak house Dylan Prime, has used RestaurantTrade since October 2000. She says she's been able to cut the time it takes to procure products and supplies in half by ordering everything online.

Even if the statistics are a bit hyperbolic, they're an apt illustration that the idea of e-marketplaces simply doesn't fit most people's notion of how to run a business.

"A marketplace is like a big bazaar or flea market when everyone goes out and shops for the best deal," says Katherine Jones, managing director with the Aberdeen Group. The problem is, that's not the way business does business. In most business transactions, there are different prices for different customers. High-volume customers get one price, while occasional customers get another. And although the e-marketplaces are more efficient than what people went through in the past to close a sale--e-mailing, calling, and meeting in person--they could never overcome perceptions that only commodities, not quality goods, could ever be bought in an open e-marketplace.

There are several other disadvantages to e-marketplaces that no one counted on. Most notably, implementing them is far more expensive than anyone ever figured, and the process isn't without its challenges. "It takes much longer and is messier to set up an e-marketplace than people thought," says IDC's Lipis. "Hooking into back-end systems of numerous companies and heavy systems integration isn't easy." And some of the expected benefits, such as lower prices, either never materialized or didn't really matter, because most large companies can already cut a percentage point or two off what they pay to their main suppliers.

What companies want more than anything is a lower cost of doing business. "The e-marketplaces that failed were built on the notion that the owners were going to get very rich very fast," says Barbara Reilly, VP and research director at Gartner. "They weren't built on solid business models, nor were they built on really making the existing business process more efficient. They never delivered on their promises."