12 months ago, B2B public exchanges were riding high as the way to conduct B2B e-commerce in the future. Specialised, all-embracing trading communities of buyers and sellers all vying to do business with one another promised low prices and low transaction costs.

Since then there have been many failures. Chemdex, a pioneering life sciences marketplace, closed last December. Earlier last year, Chemdex formed Ventro Corporation to develop B2B exchanges in other markets along the same lines as VerticalNet before it. It soon took over Promedix, aimed at the specialty medical supplies sector, but this too was forced to close with Chemdex. Petrocosm, a marketplace offering procurement tools and services for the oil and gas industry, closed in April this year. Others to suffer a similar fate include Freightwise, eGarden.com and, rather surprisingly in view of its successful B2C site, Dell Marketplace. Even sector veteran VerticalNet found it necessary to sell one of its most lucrative online marketplaces to a consortium exchange set up by leading electronics companies. What has gone wrong? Just another Internet bubble bursting? Are all such exchanges inevitably doomed?

Now the e-commerce talk is all about private exchanges. Are these just another re-hash or do they have real merit?

Vive la difference

The different types of exchanges, or marketplaces, are usually defined by their ownership. An independent exchange is owned and operated by a third party that is not considered to be a trading partner – often a B2B start-up. Ownership of a consortium-led exchange, though, is typically shared between a number of industry leaders and a technology partner. The exchange Covisint set up by General Motors, Daimler-Chrysler and Ford for the motor industry is a prime example of this type.

Confusingly, both of these types are also referred to as public exchanges to distinguish them from private ones. A private exchange is owned and operated by a single large company much in the way that EDI systems operate as hub and spokes.

The raison d’etre for the 1,500-plus third party exchanges that emerged over the past 18 months was that the Web could bring together buyers, sellers and value-added services providers (for additional services like payments, logistics, import/export, etc.) in many different industries much more efficiently than could be managed with a direct one-to-one business relationship.

Missing the point

The challenge was to entice customers and suppliers to participate in significant numbers with a substantial volume of transactions, thereby creating the liquidity and revenue these exchanges need to survive. However, two main factors have prevented their growth.

Firstly, independent exchanges have often failed to provide a compelling reason for buyers and sellers, especially the large incumbents, to participate. Not only have their prices often been no lower than dealing direct (sometimes higher, in fact), but also they have had to survive on transaction fees alone that have become increasingly wafer thin. To make matters worse, the larger and more complex the exchange becomes, not too surprisingly the higher the cost to build it. Such public exchanges often run into the $50m to $100m bracket, so how does this get paid back? In the earlier Internet years this was not a major consideration – it is now, though.

Because of the complexity of building the basic exchanges in the first place, very few of them have so far been able to introduce value-added services covering the whole transaction process, so the only way to increase fee income is through volume of transactions. However, this has led to the second factor: a backlash among the largest incumbents to form their own consortium-led exchanges, which have not fared too well either.

This is likely to get more confused still. Although consortium exchanges are generally moving forward and improving the services they offer, this may now change. Complications are arising with the establishment of the new bureaucracies to run them and with the need for integration with the diverse legacy technologies used by each consortium member. Put that together with recent concerns of anti-trust regulators and many may not make it to become much more than information hubs.

It’s relationships what count

Crucially, though, public exchanges also ignore the importance of relationships in the B2B trading process. It’s usually not just a matter of price. If it’s a question of getting a finished contract out on time to the end customer, it’s got more to do with trust, quality and performance. More than that, working efficiently with trading partners often means sharing sensitive information – something to avoid doing with competitors. Indeed, this was one of the main reasons why Volkswagen decided not to join GM/Daimler-Chrysler/Ford’s Covisint.

Public exchanges like Cordiem, which serves the aviation industry, and Covisint are effectively blocked because competitors won’t reveal proprietary information about procurement methods. Nor can exchange members be uniform in mundane-but-expensive areas such as invoicing, inventory management and documentation, foregoing even those cost savings.

IBM, on the other hand, claims specifically to have saved $1.7bn by being able to divulge sensitive price and inventory information over a private exchange built for 25,000 suppliers and customers.

These are factors that require a closer, collaborative type of relationship to get right.

Going private

In the midst of all this difficulty, the B2B focus has shifted smartly onto private exchanges. Typically, the purpose of a private exchange is to support a single enterprise trading with its own partners. The one-to-many model, but more than just a low cost EDI replacement. This collaborative process of real-time supply/demand chain management promises to utilise web-based technologies to squeeze inefficiencies out of processes across enterprises – in the same way ERP systems were designed to do for processes within the enterprise. Furthermore, this model mirrors how companies really do their business today: with a handful of partners using predetermined terms and contracts – but privately negotiated prices.

Firstly, owners of private exchanges can regulate supplier and customer access – and exclude competitors, making the sharing of sensitive information more likely.

Secondly, owners can direct suppliers and customers to use the exchange either through price incentives or by mandating changes in the way to conduct business.

Thirdly, private exchanges can be secured and tailored to serve specific projects and customers, unlike public exchanges which must be generic enough to accommodate everyone.

Some of the largest firms, for example GE, Dell Computer and Wal-Mart, have of course already successfully generated supply chain efficiencies via their private networks, so there has been little incentive for them to join the newest B2B exchanges themselves. It remains to be seen whether such companies will really find it to their advantage to relinquish direct control of their supply chains.

In the end, it looks more likely that large corporations may conduct e-commerce via proprietary networks, and come together at the consortia level to improve communications or promote common technology standards within their industries. Promoting the consortium exchange as an information portal, in fact.

Demise of the public exchange?

None of this means that the public exchange is now dead. The formula from last year has clearly not met requirements and is in need of surgery, but the private exchange alone is not really designed to serve the interests of small to medium sized companies – over 80% of the business population. In many ways, the private exchange may therefore act as a necessary building block towards fulfilling the promise of public exchanges: one-to-many begets many-to-many. The future may indeed look more like networks of private exchanges interacting with public ones. A public exchange will then need to more closely resemble an outsourced private exchange solution.

© e-principles 2001

Robin Duke-Woolley

Any comments on this article? Please send them to : Editor@e-principles.com