Let's Get Vertical

Written by Mohanbir Sawhney and Steven Kaplan

The great untold story of online commerce is that business-to-business sales have already eclipsed the higher-profile business-to-consumer market by a long shot. Annual B-to-B ecommerce is projected to soar from $43 billion in 1998 to $1 trillion by 2003, according to Forrester Research, while the consumer market swells from $7.8 billion to $108 billion in the same period. In this issue, we deconstruct the new business models — vertical, horizontal, and otherwise — that promise to turbocharge the online B-to-B engine; a handful of ambitious companies piloting those models in different industries; and the latest market data that points in new directions.

It was a busy spring for ecommerce analysts: A whole crop of numbers sprang up to put some definitive, quantitative stamp on the digital economic boom (see "B-to-Beware the Numbers," p109). The U.S. Department of Commerce weighed in with an industry summit in May on the topic of tracking ebusiness, and in June it released its second annual report on ecommerce, "Emerging Digital Economy II." Also in June, a study

released by the University of Texas — and funded by Cisco Systems — found that the Internet economy generated $301 billion in U.S. revenue in 1998, buoyed by a work force of 1.2 million people, which puts it in the same macroeconomic neighborhood as cars ($350 billion) and telecommunications ($270 billion).

While the research firms continue to run numbers, the consumer market for ebusiness continues to grab most of the headlines. Last November's declaration from Jupiter Communications that the Internet would trigger a $2.3 billion holiday shopping bonanza itself triggered an avalanche of media attention on the advent of online shopping that has hardly let up in the months since.

All that has fairly masked a quiet revolution currently under way in the digital economic realm: business-to-business (B-to-B) ecommerce. Up to this point, most of the attention in B-to-B ecommerce has focused on prominent, well-established firms such as Cisco and Dell Computer that eliminate old-economy middlemen and sell directly to business customers. But the real B-to-B ecommerce revolution is taking place outside the boundaries of individual firms.

A new breed of intermediaries is emerging to facilitate B-to-B ecommerce. These new intermediaries go by different names — "vortexes," "butterfly markets," or "net market makers." All in some way serve as electronic hubs, each spinning in a new market. These hubs focus on specific industry verticals or specific business processes (from spare airplane parts to secondary mortgages), host electronic marketplaces, and use various market-making mechanisms to mediate any-to-any transactions among businesses. They create value by aggregating buyers and sellers, creating marketplace liquidity (a critical mass of buyers and sellers), and reducing transaction costs.

What makes any of this revolutionary? First, these intermediaries do for ecommerce transactions what a network hub does for bits: concentrating, routing, and switching transactional traffic in B-to-B ecommerce. Second, they occupy a central position between buyers and sellers, much as airline hubs do between city pairs.

In short, hubs promise to reshape the landscape of B-to-B ecommerce — and here are a few of the lesser-known "numbers" that back it up. First and foremost is the Forrester Research projection that B-to-B ecommerce will surge past $1 trillion by 2003. San Francisco-based investment bank Volpe Brown Whelan estimates that overall hub revenue (both transactions and advertising) nationwide will grow from $290 million in 1998 to $20 billion by 2002. The Precursor Group's estimates are even higher, putting B-to-B exchange revenue between $50 billion and $130 billion by 2002. Even with the more conservative estimates, hubs could generate transaction fees of more than $10 billion within three years, with gross margins of 85 percent. If these numbers sound too optimistic, consider eBay, a hub in the consumer-to-consumer market. eBay boasts gross margins in excess of 80 percent and, more strikingly for an Internet company, reported a profit within a year of its inception.

Despite their enormous significance for B-to-B ecommerce, hubs remain poorly understood. Like the bulk of an iceberg that lurks below the waterline, they remain largely invisible to the media, investors, and analysts. While the landscape is still blurry,

and few hubs have achieved any degree of prominence, it is possible to describe what hubs are, what they do, how they create value, and what the future is likely to hold.

B2B: A Different Animal

Returns to scaleThis is the most important and perhaps the least understood difference between consumer commerce ventures and B-to-B hubs. Consumer hubs, even as expansive as Buy.com's, are one-way networks that deal directly with buyers and create benefits mostly for sellers. B-to-B commerce hubs tend to be two-way networks that mediate between buyers and sellers, and create benefits for both sides. Lots of benefits: The value created by consumer hubs tends to increase linearly in the number of buyers; the value created by B-to-B hubs increases as the square of the number of participants.

Here's the math: At Buy.com, the benefits to an individual buyer are roughly the same whether there are 100 customers or 100 million customers, because the benefits to buyers are primarily the time saved from going to a physical store and looking for an item. Buy.com does, however, benefit on the supply side, by tallying savings in marketing and procurement. These benefits tend to be linear, so the total value created by Buy.com increases only linearly in the number of customers.

Now, consider a B-to-B hub: It creates value by reducing search costs, reducing information transfer costs, standardizing systems, and improving matching for both buyers and sellers. Buyers benefit because they have more choices and sellers benefit because they have access to more buyers. All of these value drivers increase with the square of the number of participants in the hub.

Take the case of five potential sellers and buyers in a B-to-B market. In the absence of the hub, each seller would have to determine the identity of each buyer, through advertising or a direct sales force. Each seller would have to contact each buyer each time it wanted to do a transaction. Without a hub, the participants would have to undertake 25 searches — each seller looking for five buyers — and then make 25 contacts (either faxes or phone calls) each time the sellers wanted to sell.

Now look at this electronically: the hub finds the buyers and sellers, reducing the number of searches to 10. Similarly, each time the sellers want to sell, there are only 10 contacts — five postings on the hub, and five views by the buyers.

The methodology is similar for matching buyers and sellers, standardizing systems, and transferring information such as credit checks, product descriptions, and evaluations. Further, the complexity of the benefits a hub offers makes it difficult for competitors to offer customers similar benefits. This is particularly true for matching between buyers and sellers in auctions and exchanges. A buyer is far less likely to find a match in an illiquid hub than in a liquid one. Small wonder, then, that eBay is profitable and commands a rich valuation.

Importance of domain expertiseSetting up shop as an online retailer does not require deep knowledge of specific

categories. The founders of many popular online shops had no previous category experience, and Amazon has quite easily migrated across retail categories. In contrast, domain expertise is a must for creating a hub. Consider a hub like SciQuest.com that mediates between buyers and sellers in the laboratory and scientific equipment marketplace, or PlasticsNet.com (see "The Trading Post," p102) that has spawned a new market between plastics manufacturers and plastics processors, or e-Steel, that makes a market between buyers and sellers of steel. The founders of these companies had extensive industry experience and relationships with key buyers and suppliers. Domain expertise and relationships are key barriers to entry for hubs.

Customer acquisition and retentionOnline retailers typically use advertising and affiliate programs for customer acquisition. But B-to-B buyers and sellers don't simply see a banner advertisement and sign up with SciQuest.com or PlasticsNet.com. Customer acquisition requires sales calls, and the process for signing up buyers and sellers is time-consuming and expensive — supplier catalogs have to be loaded online (see "Chemdex's New Supply-Side Economics," March '99, p47), business processes need to be understood, business rules need to be defined, and the hub's systems need to be integrated with those of its buyers and sellers. Customer switching costs and customer retention rates are also correspondingly higher for hubs, once they embed themselves into the business processes of buyers and sellers, and for them, competition is a lot further away than a mouse click.

All that presents a worthwhile industry-wide tradeoff: The higher entry barriers due to increasing returns to scale, domain expertise, and higher customer switching costs add up to stronger profit potential and more defensible business designs than B-to-C retailers.

B2B Hubs: What's What

In contrast to pure financial marketplaces, hubs are contextual marketplaces; hubs focus on a specific dimension of it. Attempting to be everything to everybody is a recipe for failure. Nets Inc. was designed as a B-to-B shopping mall across different verticals and different functions. One of the primary reasons it failed is that it had no focus or context. It was neither vertical nor functional, and never able to attract enough buyers and sellers to generate liquidity.

A hub, though, can specialize vertically along a specific industry or market, or it can specialize horizontally along a specific function or business process. Based on these dimensions, the universe of hubs boils down to two primary types: vertical and functional. Together, they form the quilt of B-to-B ecommerce.

Vertical hubsVertical hubs serve a vertical market or industry focus. They provide deep domain-specific content and domain-specific relationships. Examples: Altra Energy (energy), Band-X (telecommunications), Cattle Offerings Worldwide (beef and dairy), SciQuest.com (life sciences), e-Steel (steel), Floraplex (florists), IMX Exchange (mortgages), PaperExchange (paper), PlasticsNet.com (plastics), and Ultraprise (secondary mortgage exchange). Vertical hubs typically start out by automating and

hosting the procurement process for a vertical, and then supplement their offerings with industry-specific content.

The likely success of a vertical hub increases with:

Greater fragmentation among buyers and sellers.

Greater inefficiency in the existing supply chain.

Creating critical mass of key suppliers and buyers.

Domain knowledge and industry relationships.

Creating master catalogs and sophisticated searching.

Adjacent verticals for leveraging existing supplier or buyer base.

The primary challenge for vertical hubs is the difficulty of diversifying and extending their business into other vertical markets, because their expertise and relationships are fairly domain-specific.

Functional hubsFunctional hubs focus on providing the same functions or automating the same business process across different industries. Their expertise usually lies in a business process that is fairly horizontal, which means that it is scalable across vertical markets. iMark.com, for example, focuses on buying and selling used capital equipment. Its target participants are investment-recovery managers responsible for the equipment. Other examples of functional hubs include Processors Unlimited (reverse logistics), MRO.com (maintenance, repair, and operating procurement), Employease (employee benefits administration), Celarix (global logistics monitoring and tracking), BidCom (project management), Adauction (media buying), and YOUtilities (energy management).

The likely success of a functional hub increases with:

Degree of process standardization.

Process knowledge and work-flow automation expertise.

Complementing process automation with deep content.

Ability to customize the business process to respond to industry-specific differences.

The primary challenge for functional hubs is to deliver industry-specific content. They target functional managers who affiliate and organize their work primarily around their functional area, and not their industry. But many functional managers also affiliate with their industry. The risk: They will gravitate toward a vertical hub for their industry and relegate the functional hub to become a back-end service provider for the vertical hub.

Separating the Winners & Losers

Architects of B-to-B hubs must address a number of key design and execution issues. These include:

Choosing a marketmakerHubs can employ a variety of market-making mechanisms to mediate transactions between participants in the hub. These mechanisms can be fixed-price mechanisms that are typical of catalog purchasing, or dynamic pricing mechanisms that include auctions, exchanges, or barter.

A listing or catalog model creates value by aggregating suppliers and buyers. It works best in industries characterized by fragmented buyers and sellers who transact frequently for relatively small-ticket items. Given the small transaction size, it is too costly, even on the Net, to negotiate each transaction. The catalog model also works well when most purchasing takes place with prequalified suppliers and with predefined business rules, and the occasional purchase requires searching across a number of smaller suppliers. Finally, it works best for situations where demand is predictable, and prices do not fluctuate too frequently. Chemdex, SciQuest.com, and MRO.com are examples of catalog-centric hubs.

Auction models create value by spatial matching of buyers and sellers. They work best in industries or settings where one-of-a-kind, non-standard, or perishable products or services need to be bought or sold among businesses that have very different perceptions of value for the product. Capital equipment, used products, unsaleable returned products, and hard-to-find products fit this description. iMark.com uses an auction model to sell used capital equipment. Adauction.com auctions off perishable online and print

advertising inventory.

Exchange models create value by temporal matching of supply and demand (see "The Matchmaker," p96). They require a real-

time, bid-ask matching process, marketwide price determination, as well as a settlement and clearing mechanism. The exchange model works best for near-commodity items that can have several attributes, but are easy to specify. Exchanges create significant value in markets where demand and prices are volatile by allowing businesses to manage excess supply and peak-load demand. PaperExchange in paper, e-Steel in steel, and Altra in energy are all good examples of hubs that employ exchange models.

Barter models create value by matching two parties that possess reciprocal assets within an asset class or across asset classes. While barter has traditionally been used in inflationary economies with shortages of hard currency as a tool to minimize currency risk, there are other innovative applications, such as bartering manufacturing capacity, bartering services for other services, and bartering high-transportation cost assets (such as paper or steel).

In many cases, a hub will find it valuable to offer more than one market-making mechanism. Customers of hubs will favor hubs that allow buyers and sellers to choose the appropriate market-making mechanism. This means that hub architects need to take care in choosing a technology platform, because technology providers have tended to

focus on catalog (Ariba, Commerce One, IBM), auction (Moai Technologies, Dynamic Trade, OpenSite Technologies), or exchange (Tradex Technologies) models, and lack integrated multimechanism platforms.

Solving the chicken-and-egg dilemmaThe value created by a hub increases non-linearly in the number of participants. The key goal of any hub, therefore, is to attain liquidity as quickly as possible. The problem, of course, is that it is difficult to attract buyers without sellers and difficult to attract sellers without buyers.

While hubs need to market to both sellers and buyers, they will generally be better off marketing more heavily to the party in the transaction that receives the most benefit. Once the hub has gained the participation of that side, it can market more easily or even compel the other side to join up. Despite conventional wisdom that hubs tend to be buyer-centric and hence need to attract buyers first, there is no reason why hubs will consistently favor buyers over sellers or vice versa. Chemdex and SciQuest.com focused on buyers first in order to attract sellers. PaperExchange, in contrast, has focused on sellers first in order to attract buyers.

In its early stages, hubs might also consider injecting liquidity into the market. Consider the strategy that priceline.com adopted in growing its market for airline tickets. It focused on getting buyers, who arguably receive the greater relative benefit, injecting liquidity by buying cut-rate airline tickets in order to attract buyers. Once it achieved critical mass, the incentives for airlines to cooperate increased.

How Does Your Merchant Grow?   CATALOG

(Chemdex, PlasticsNet.com, SciQuest.com)

AUCTION(AdAuction, IMark.com)

EXCHANGE(e-Steel, PaperExchange, Ultraprise)

How it works

Demand/supply aggregation

Spatial matching Temporary matching

How buyers benefit

Lower search and transaction costs; broader supply base

Catalog benefits, plus better matches, better prices

Auction benefits; peak-load demand management; hedge risk in volatile markets

How sellers benefit

Broader customer access, lower transaction costs

Catalog benefits, plus better pricing

Auction benefits; liquidate excess supply; manage volatility

Where it works best

MRO products; pre-planned purchases; fragmented supplier base

Used capital equipment; perishable capacity; hard-to-specify products

Near-commmodities; high-fixed cost assets; volatile markets

How prices are set

Pre-negotiated, usually static

Most attractive bid, prices move in one direction

Marketwide bid-ask; moves up and down

Can buyers be sellers?

No Sometimes Yes

Key challenges

Creating master catalog; gaining supplier critical mass

Liquidity, misrepresentation/fraud, fulfillment

Asset specificity; off-exchange trade

Timing market entry How quickly should a hub open for business? Let's overstate the obvious: Those that start early reach buyers and sellers more quickly, preempt competitors, and begin to learn earlier about the market. But these benefits must be weighed against the risk that premature opening will discourage market participants from returning because of lack of functionality or liquidity and alert potential competitors. Early entry, then, makes good sense. Another strategy can be to open an informational or content-focused hub, and to add transactional functionality as liquidity improves (see "The Industrial-Strength Portal," p118).

Managing channel conflictSome existing intermediaries are initially likely to be hostile to hubs, because of concerns around disintermediation and price erosion. Hub architects should stress that it can complement an intermediary rather than act as a substitute. Hubs can provide more volume and better matches to existing intermediaries. This is particularly useful in industries with uncertain or volatile supply and demand. To counter price erosion, sophisticated hubs should provide value proposition transparency, not merely price transparency. How? By offering metrics related to quality, reputation, reliability, speed, or service, in addition to providing prices. Finally, hubs can create "virtual private marketplaces" that preserve prenegotiated terms and relationships between specific buyers and suppliers.

Expanding the scope of the offeringWhile liquidity is the key determinant of a hub's success, those firms should also try to increase the depth and breadth of its relationships with participants. This can be done by providing complementary services to participants that make it more costly for buyers and sellers to transact elsewhere. Such services might include IT services like system integration and hosting; financial services such as payment processing, receivables management, and credit analysis; logistics services like shipping, warehousing, and inspection; and risk mitigation services like escrow and warranties. Partners such as Skyway (supply-chain management), PaylinX (enterprise payment servers), i-Escrow (escrow), eCredit.com (credit analysis), and USinternetworking (application service provider) help to round out the offering. In addition, each of these complementors are potential sources of referral revenue from the existing participant base.

Managing growth and diversificationOver time, the hub will need to diversify beyond its initial choices of strategic position. These growth vectors can be along four key dimensions — horizontal scope, vertical scope, offering scope, and mechanism scope. Consider e-Steel, a vertical hub specializing in the steel industry, the procurement process, using an exchange mechanism, and by

largely outsourcing logistics services. Horizontally, e-Steel could add logistics tracking or investment recovery services. Vertically, it could branch into the packaging vertical. To expand scope, it could offer credit analysis or fulfillment services. Choices should be based on the strength of connection to the new supplier and buyer base, the new business process, the knowledge and relationships needed, and the alternative market-making mechanisms within the current participant base. In some cases, partnerships and acquisitions will drive growth.

Forecasting the evolution of B-to-B hubs

A few predictions of what the next year or two holds for hubs:

Hubs will have winner-take-all characteristics: The strong increasing returns characteristics will create even more scale advantages for the first hub to achieve scale and liquidity than in the consumer portal or retail business. Even being the second-biggest player may not be enough.

Vertical hubs will find it hard to diversify beyond their verticals: They compete on domain-specific relationships and expertise. Unless they can find closely related domains where they can leverage these assets, they will find it difficult to diversify into other vertical markets.

Vertical hubs will form a patchwork of alliances with functional hubs: Verticals possess domain expertise but lack functional expertise, while functional hubs possess functional expertise, but lack domain expertise. Alliances will form across the "quilt" between verticals and functionals. In these alliances, vertical hubs will usually emerge as the ones that control the customer relationship.

Software vendors will climb out of their silos: Currently, software vendors sit in three silos that correspond to their market-making mechanism of choice — transaction software vendors (such as Ariba and Commerce One), auction software vendors (such as Moai and OpenSite), and exchange platform vendors (such as Tradex; see "The Enabler," p112). The walls of these silos will break down, and a flurry of mergers and alliances between software vendors will ensue.

Exchange models will evolve to include derivatives: The exchange mechanisms in hubs currently are limited to spot markets. As participants become more sophisticated and hub software platforms improve in functionality, hubs will begin to offer derivative products like forward contracts and options on commodities and manufacturing capacity.

All except the biggest firms will give up on hosting hubs: Early generations of B-to-B ecommerce software focused on catalogs, auctions, and exchanges hosted by individual firms. This firm-centric model limits liquidity, and will give way to catalogs hosted by hubs.

New "metahubs" may emerge with shared infrastructure and services: Although vertical hubs will not consolidate across vertical domains, there is no reason for them to have dedicated infrastructure and supporting services. We may see the

emergence of new market hubs with shared back-ends and common functional hubs servicing the different "tenant" vertical hubs.

The power of conventional commodity exchanges will erode. Conventional commodity exchanges are devoid of context, they will find it difficult to compete with the powerful integration of context and exchange functionalities that hubs will provide. They also lack business process integration capabilities. As a result, hubs will gradually drain liquidity from conventional commodity exchanges.

Market analysis at this stage — like any analysis about the future of ecommerce — boils down to advanced guesswork. But if the recent successful IPOs of online B-to-B companies such as VerticalNet and Ariba are any indication, and if the market projections are anywhere close, the best is clearly yet to come.

Mohanbir Sawhney (mohans@nwu.edu) is the Tribune Professor of Electronic Commerce and Technology at the Kellogg Graduate School of Management, Northwestern University, and heads the ecommerce and technology group.

Steven Kaplan (steven.kaplan@gsbpop.uchicago.edu) is the Neubauer Family Professor of Entrepreneurship and Finance at the University of Chicago Graduate School of Business, and the faculty director of the entrepreneurship program.

Ultraprise: The Matchmaker

How an electronic go-between is rewiring the mortgage market.

By Mark Halper

Until recently, the now-defunct Smitty's Pool Hall on German Street in Shepherdstown, W.Va. (population 1,287), hosted the state's annual arm wrestling championship, a bellicose event in which the best display of high technology was the art of launching humans through the tavern's windows — or so say the old-timers.

Hardly the typical locale for a promising Internet startup, right? "It's a weird place," allows David Levine, the 33-year-old co-founder and CEO of Ultraprise, a new online business-to-business broker in the $1.5 trillion mortgage industry. But not so weird that state development funds couldn't induce him to refurbish Smitty's three years ago and wire it up into a joint now jumpin' with computer servers and software programmers dead set on adding efficiencies to the cumbersome process of bank-to-bank mortgage sales.

Levine, who sometimes commutes to work by paddling a canoe down the Potomac River, is riding a new current of information flow that could change the way banks resell mortgage loans. Ultraprise thrives as an electronic go-between for banks that originate mortgages and banks that buy them — a lively market known as the secondary mortgage market.

Invisible to most of the consuming public, the basic business works like this: Mortgage originators may hold onto loans or sell them into the secondary mortgage market. Roughly 62 percent of mortgages originated are sold into the secondary market. Mortgage bankers commonly package their loans into financial instruments called mortgage-backed securities. These are sold to investors such as mutual funds, pension funds, and depository institutions, which profit on the pooled interest payments.

Originators typically will not lend money to a consumer unless they know they can turn around and sell the loan, a step that effectively acts as a guarantee for them. Levine's startup matches buyers and sellers of "nonconforming" loans — those that the federal government's Fannie Mae and Freddie Mac enterprises decline to guarantee and buy from originators. (Fannie and Freddie only buy loans that "conform" to an ever-changing definition of safe loans. In mortgage parlance, neither will back "jumbo" loans, or those presently more than $240,000, and Fannie Mae will not buy "sub-prime" loans, which are loans made to consumers with questionable credit ratings.)

Still, many buying banks are willing to take on the risk of these loans. Of the $1.5 trillion in mortgage loans originated last year, $480 billion, or about a third, sold as nonconformers, according to the Mortgage Bankers Association of America.

As with infomediaries in other markets, Ultraprise offers the considerable advantage of quick turnaround. Levine claims that conventional bank-to-bank mortgage trades take several weeks to process, while his system reduces it to a few hours. The upshot? An

originator can close a consumer loan more quickly (originators often do not make a loan until they know they have a buyer), a key selling point in the competitive home-loan business.

Ultraprise's enhancements to the marketplace reach beyond the simple joys of saving time. The company provides information that helps buyers find the type of loan they are most interested in purchasing. Some buyers, for instance, may want to buy a package of loans made to particularly risky credit ratings, because these loans carry higher interest rates and therefore represent higher potential profits. These and other buyers may also want to avoid making loans to consumers who have a tendency to pay off loans early, because early payers deprive financial institutions of long-term interest payments. Levine's system also taps information that shows likelihood of an early payoff.

A nonconformist marketHow does this budding online marketplace function? On the Ultraprise site, sellers post information that helps buyers analyze the makeup of the loan packages. If one of the loans in a pool of, say, 10 loans falls wildly outside the buyer's parameters, the buyer may request that it be excluded.

All that information, in turn, feeds into the supply and demand equation that could net the seller a fatter price of, say, $535,000 instead of $510,000 on a $500,000 loan. Notes Dave Tabors, a venture capitalist with Wellesley, Mass.-based Battery Ventures, "an originator may not get a good price if there are only three or four buyers. It may turn out that buyer number 20 has a real need and is willing to pay a premium." Battery has invested $5 million in Pedestal Capital, another online secondary-mortgage broker and a direct competitor to Ultraprise.

Pedestal charges sellers for its services, so naturally Tabors promotes seller advantages. Kevin Jones, editor of the newsletter Net Market Makers, compares Pedestal's business to one of the consumer world's best-known online infomediaries, Priceline.com, the state-your-price travel- and auto-buying site with a $14 billion valuation. Priceline calls its patented pricing model "demand collection." The same concept applies to Pedestal, Jones says, although Pedestal does not use a state-your-price mechanism. "In any inefficient market, you can find sufficient demand to create efficient markets," says Jones.

Levine agrees that the non-conforming secondary mortgage industry could use some efficiencies. Sellers have fewer buyers to choose from, which undermines potential profits; for buyers, a limited source of supply threatens the balance of their portfolios. And since real estate markets vary widely across geographic regions, a heavy concentration of loans in one area can sting in bad times. (A drug bust in Florida's Dade County several years ago, for instance, devastated the mortgage holdings of several banks focused there, as defaults mounted on jumbo loans.)

Levine has been busy tinkering with a different revenue model than Pedestal's; when Ultraprise launched its exchange last February, it was charging sellers, but business got off to a slow enough start that by June the company reversed course and shifted to charging buyers — "who were paying for it anyway," Levine quips, in the form of charge-backs by the seller. At the same time, Ultraprise replaced its percentage-based

fees with a flat fee of $75 per sale on home-buyer loans and $60 on second mortgages. Under the old structure, commission fees ranged from 0.1 percent to 0.3 percent, so that "for a certain type of pool" they were prohibitive, says Levine.

Membership fees have also been overhauled. In June, Ultraprise began waiving its $5,000 annual membership charge if a member trades more than $100 million worth of loans in a year. The company also waived a $25,000 signup fee through the end of the year.

The June changes seem to have picked up activity; Levine says that Ultraprise that month added 32 members to the 15 it already had, and that he hopes to have more than 150 by the end of the year. Members include HomeSide Lending of Jacksonville, Fla., and GMAC-RFC in Minneapolis, which ranked 8th and 16th respectively in first-quarter mortgage originations, with volumes of $7.86 billion and $4.37 billion, according to Inside Mortgage Finance. Others include Telebanc Financial, the Arlington, Va.-based online bank acquired in June by E*Trade; PSB Lending in Carlsbad, Calif.; and Boca Raton, Fla.-based Westmark Mortgage, a leading buyer of sub-prime loans.

So convinced are some members, Levine says, that they're preparing to jointly make a $15 million investment in his company. (Midsummer, he was closing a second round of venture financing at $9 million, including $1.5 million from first-round investors West Ven Partners of West Virginia, and angel investor and Trusted Information Systems founder Steve Walker. A more modest first round brought in $1.1 million.)

All that's not to say that the Mountain State-based company will find its climb to the top unobstructed. Ultraprise faces a formidable challenge from Pedestal, which launched its exchange in early 1998 and which, unlike Ultraprise, does not charge membership fees, waived or otherwise. Pedestal backer Tabors argues that such fees pose unnecessary barriers, especially for mortgage originators whose main goal is to keep a liquid source of funds to help them make their next loan. With fees, he says, "People might say, do I even want to try it?"

If Ultraprise and Pedestal have a common enemy, it is the network of industry cronies that observers say influence the way business is conducted. The slickest, most streamlined electronic deal-brokering system may have little impact if time-honored mortgage brokers ranging from local five-person shops to larger companies like Sandler O'Neill in New York City, Spires Financial in Houston, and Pamex Capital Partners in Edison, N.J., don't play along.

Wall Street dealers also get a hand in the action sometimes by going straight to the originator rather than waiting to buy pools of loans from banks that buy from originators. Pedestal's founder Yung Lim estimates that 50 percent of all secondary mortgage sales today go through traditional brokers and dealers.

With the importance of cronyism in mind, both companies have carefully hired individuals from inside the industry. "We just decided we'll hire everybody's friends and conquer the market," says Levine, whose head count now stands at 90, or six times Pedestal's head count of 15. Not a bad idea since just five years ago Levine was writing reports about African agriculture and Brazilian education for the World Bank.

Ultraprise last spring landed Dave Matthews as its new president and chief operating officer. Matthews, 50, had been chief information officer, overseeing a staff of 500 at Chase Manhattan Bank, the leading mortgage originator in the country in the first quarter of this year. He also has served on committees and advisory boards with Fannie Mae and the Mortgage Bankers Association.

Competitor Pedestal has made similar moves: Over the summer, it named Stuart McFarland president and CEO. He previously held senior executive positions with GE Capital and Fannie Mae. Pedestal founder Lim himself spent six years at Merrill Lynch, where he was a vice president of risk management in the mortgage department.

Given all that, the question remains: Is there a need for a new Web-enabled middleman, when banks have been finding each other without them for years, both directly and through physical-world brokers?

"I can't say you hear a lot of complaints from participants in the market that they are not able to get lots of bids," says Inside Mortgage Finance managing editor John Bancroft. "If they [Ultraprise] show they add value, they'll be rewarded."

PlasticsNet: The Trading Post

Taking on a 92 billion-pound gorilla — one buyer at a time.

By Jeffrey Davis

It's been more than three decades since that smug, middle-aged authority figure in 1967's The Graduate offered Dustin Hoffman's character some famously obtuse career guidance: "I just want to say one word to you — just one word," he advised. "Plastics!"

The joke was on the grownups, of course, seen as hopelessly out of touch, still clinging to fading blue-sky dreams of the industrial age and the Company Man. Everyone could smirk — everyone except, maybe, the hard-working folks across America punching timecards in research laboratories under smokestacks, helping to bring forth such 20th-century marvels in polymers as polypropylene and polyurethane. (Folks you might like to thank, actually, the next time you pull on a GORE-TEX parka on the ski slopes.)

Flash forward to 1999: The tables have turned a bit, and the plastics industry is stepping into a different scene altogether. Exit the industrial age, enter the information one. Exit middle-aged authority figure; enter high-rolling venture capitalist who was probably in diapers when The Graduate hit theaters. Exit Dustin Hoffman, enter brothers Tim and Nick Stojka, recent heirs to a family-owned plastics-supply business outside Chicago — and the hard-driving co-founders of the burgeoning new B-to-B trading hub for the plastics industry, called PlasticsNet. What is the venture capitalist whispering to the entrepreneur? "I just want to say four words to you about plastics — just four words: Business-to-business infomediary."

Like many of the emerging ecommerce players in business-to-business markets, 4-year-old PlasticsNet sits on a titanic ebusiness opportunity — big enough, some believe, to make Jeff Bezos and all the goodies he can stuff into Amazon's closet look like a tinker-toy business model in comparison.

Web entrepreneurs might still be more easily drawn to building digital storefronts for gifts and flowers than they are for facilitating shipments of, say, 3 tons of high-grade ethylene-butyl acrylate. But all career jokes aside, "plastics" is standing pretty tall at the end of this century. The plastics business is currently a $370 billion-per-year enterprise, has been growing at a 55 percent clip since 1991, and represents the fourth-largest manufacturing segment in the United States. Multinational firms such as DuPont and GE Plastics dominate the playing field. Together, it produces nearly 92 billion pounds per year of 30,000 different grades of materials and fibers — from basic resins to plastic part molds — that wind up in everything from your PC's mouse ball to your kid's blow-up wading pool.

The market's many disparate and entrenched players range far and wide: More than 5,000 plastics suppliers cater to more than 18,000 plastics processors, and account for $85 billion in revenues for raw materials, equipment, and other products. Wedged in between are more layers of distributors (who manage inventory for geographic regions), drawing

another $5 billion out of the market; small-time, phone-and-fax plastics brokers who unload "odd lot" or "off spec" shipments of plastics, but still generate big-time ($500 million) sales; maintenance, repair, and operations (MRO) companies that account for another $28 billion in plastics-related business; and more than 300 "compounders," companies that produce customized, low-volume, high-margin specialty materials. All together, a sticky business — a 92 billion-pound gorilla that no one in their right mind would think about re-engineering overnight.

Traffic first, transact laterThat is, until the Stojkas came along. In 1994, Tim Stojka and brother Nick were picking up where Dad's retirement left them off, running the family business called Fast Heat, a Chicago-area manufacturer of heating and control systems used in plastics processing. With one eye on the emergence of the Web, and another on the frustrating inefficiencies of their own business — cumbersome, greedy middlemen such as distributors who rung up 30- to 50-percent commissions on their products — the brothers Stojka decided not just to get mad; with visions of coaxing that 92 billion-pound gorilla onto the information superhighway, they just got online.

At an industry trade show in September 1995, they launched the Plastics Network, a community-based Website where buyers and sellers could find each other electronically for the first time. The concept grew legs: What began as a bulletin board service gathered its own inertia through visitor traffic, and eventually added technical forums and classifieds, and requests for quotes and information.

Relaunched earlier this year by the Stojkas' privately held ecommerce parent company Commerx (with seed funding from San Francisco-based Internet Capital Group and others), the Plastics Network is now a full-fledged ecommerce engine renamed PlasticsNet.com, with 52 full-time employees — expected to balloon to 100 by 2000. Today the site sees more than 90,000 visitors a month, has signed up more than 200 sellers — who pay a $5,000 to $8,000 annual fee to have their storefronts and catalogs posted on PlasticsNet — just introduced an online exchange for resale materials, and has plans to partner with leading online back-end logistics provider J.D. Edwards to streamline an increasing load of transactions.

The model is "classic infomediary," says Charlie Finnie, a leading B-to-B ecommerce analyst and managing director of San Francisco-based investment bankers Volpe Brown Whelan. Commerx takes a 5 percent to 10 percent cut of transactions, undercutting an army of land-based brokers and distributors. The result: The company's current ad-dominated revenue model is expected to graduate to transaction revenue by mid-2000. Indeed, sales jumped 300 percent in 1998, and are now on track to increase another 500 percent in 1999. Though the company won't divulge specific sales figures, it is not yet profitable, but expects to be in the black sometime next year. As Finnie points out, "Like a lot of these new trusted third-party [Internet] businesses, if PlasticsNet just gets to a point where they get 1 percent market share, that's huge. That's a billion-dollar business. I'd sink some money into that, wouldn't you?" In fact, the company hopes to grab 5 to 10 percent of the market.

Buy-side biasA quick glance at PlasticsNet's brief history — launching an ecommerce holding company back in '95, turning down a $10 million offer for the Website the following year — makes it seem like Tim Stojka, co-founder of both Commerx and PlasticsNet (he is also CEO of Commerx and president of PlasticsNet), has been eerily prescient about the future of B-to-B ecommerce. But he makes no such claim, insisting that PlasticsNet's early success has come from building it around the needs of the buyers first, sellers later.

"Everybody is calling these things vertical markets, but we're just focusing on a buyer segment — plastics converters, companies who need and buy raw materials equipment, process it, and produce some sort of end product," says Nick Stojka, the company's co-founder and executive vice president. "Unlike the other verticals, we're uniquely focused on the buyer.... We're not just providing them with the raw materials, we're providing everything they buy on a regular basis, from a light bulb to spare parts...even the additives or resins."

Along with 30,000 grades of materials, plastics buyers need a tailor-marketed slew of expensive machinery — blenders, feeders, heaters, loaders, granulators, pulverizers, and the like — to get their work done. Each of those are high-ticket ($10,000 to $15,000 is typical), high-margin items on which Stojka has his eyes.

But long before a buyer comes to PlasticsNet in search of anything, the site boasts community services to lock them in for many returns. Visitors sign up free to become PlasticsNet members. What do they get for their eyeballs? Here's a brief inventory:

An expansive inventory of technical datasheets on plastics products and equipment.

Supplier Websites and searchable catalogs.

The largest online plastics-industry job bank. (Within a week of its launch in May, it had 400 listings.)

An education database and 12 heavily trafficked technical forums — from "blow molding" to "thermoforming" — that roam the minutiae of plastics processing.

Six sections of classifieds postings, including project bids and listings for used materials.

"It's a different strategy than a lot of portals," says Cary Weldy, Commerx's 33-year-old director of strategy development who spearheaded the ecommerce launch of PlasticsNet. Like the Stojkas, Weldy is no wet-behind-the-ears Java hotshot flown in from Silicon Valley. He's another son of a plastics-business father, and a former marketing manager at General Polymers, one of the biggest plastics suppliers in the country. "Anybody can launch an auction and call it whatever.com, but what makes something long-term sustainable is the ability to develop community."

While the Stojkas stick to their buy-side focus to gather that coveted critical mass of market players, some on the supply side aren't exactly holding their breath in anticipation.

Barbara Castilano, marketing consultant for Dayton, Ohio-based Comet Automation Systems — a dues-paying PlasticsNet supplier — is in the niche business of selling such critical plastic-biz gizmos as the "Exact-A-Batch Gravimetric Blender." At $10,500 apiece, the Gravimetric is "never going to sell online," she says. It's a configurable, customizable piece of machinery that buyers need to check out for themselves. But she's sold at least on PlasticsNet's ability to be a solid, cost-effective sales-lead generator — another first-stage function for B-to-B infomediaries. Comet's listing on PlasticsNet, she says, draws a minimum of 20 sales leads a month. "[PlasticsNet] is one of the few search engines that you have to pay for, and generally I'm opposed to that. But the leads have been solid. This one's proven it's cost effective."

The master planWhile VerticalNet (see "Industrial-Strength Portal," p118) is test-driving a different model — claiming ownership of nearly 50 vertical markets, and scrambling to fill each of them with content, community, and eventually transactions — the Stojkas and Weldy are content to "go deep" in every aspect of plastics buying and selling alone, gradually building out buyer-centric services one at a time, from full-service procurement to "helping them do forecasting and demand planning," Nick Stojka says.

After that? That's the card Weldy and the Stojkas are holding face-down at the moment, but the implied plan is clear: If they can prove they can master the complexities of the plastic business online — yes, an IPO looms — Commerx will move to other "similar industry-based" markets to test the true potential of B-to-B infomediary theory even further.

Prepping the next billion-pound gorilla market in manufacturing for streamlined Web commerce is clearly part of Commerx's master plan. But first, as Weldy hammers again and again, "We're not trying to be a jack of all trades. We need to be the master of one first."

B-to-B, By the Numbers

By Kim Cross

As the commercial Internet evolves from a speculative, capital-crazy casino into a truly measurable economic phenomenon, the folks in charge of the numbers — the accounting and research firms such as Forrester Research, Jupiter Communications, GartnerGroup, and others — no doubt have all the work they can handle for years to come. Yet today, trying to make sense of the market projections — with Forrester's $1.3 trillion estimate for 2003 B-to-B ecommerce leading the pack — is a little like what Alice saw through the Looking Glass: strange, new wonders, but only half-real.

"[Research] is as much an art as a science," says Geoffrey Ramsey, statsmaster for eMarketer, a company that aggregates and analyzes statistics from many research houses. "There's no such thing as perfect research. By looking at a number of different sources for a topic, you'll get a better picture of reality."

Discrepancies arise from variables that drive the numbers in different directions. Each company designs its own research architecture and researchers have different parameters for defining concepts such as ecommerce. Some count only those purchases transacted online. Others incorporate sales "facilitated" online, which could include items researched online and purchased elsewhere. Some have looser definitions still, including online advertising sales in the revenue figures.

Then there are many variables within a study's methodology. The size and demographics of the sample surveyed (dozens of CEOs vs. thousands of consumers), the nature of interviews (qualitative vs. quantitative), and polling devices (Internet questionnaires vs. face-to-face interviews) are a few examples.

Online surveys are also prone to a self-selection bias, misrepresenting the population at large by eliminating those who are not online or don't have time to complete a survey — often the very people relevant to the survey.

The basis for extrapolation can also cause results to vary widely. Some researchers base their projections on estimated demand, asking customers to approximate their past spending and predict what they'll spend in the future, while others use supply-side figures, using expected sales as estimated by the vendors themselves.

Finally, there is old-fashioned bias. "There's room for interpretation and subjective guesswork," Ramsey says. In an economy like this, it's hard not to be optimistic when calculating projections. Moreover, a circle of influence created by the companies that supply research firms with the data they later purchase.

"When you are selling [market reports] to the very same people you're getting the data from there's going to be a problem," Ramsey warns.

A word to the wise: heed as many sources as possible and draw your own careful conclusions.

Tradex: The Enabler

Building the platforms for online B-to-B.

By Susan Moran

Horse feed. Plastics. Computer chips. Name any market up for grabs in the rapidly expanding realm of online business-to-business, and Tradex Technologies aims to build the digital engines to bring them up to speed.

Every one of these complex new electronic marketplaces, from plastics to paper, demands a set of sophisticated tools that can cater to the needs of thousands of buyers and sellers who converge in one meeting place at one time to do their business. The tools of the online B-to-B trade today generally include a sophisticated transaction engine with a user interface, an electronic catalog of some kind, software that enables both the purchasing and selling of goods, configuration tools, and the like. Some vendors have lined up on the sales side of the commerce equation, while others offer tools just for buyers. Tradex serves as a puppeteer of sorts, offering technology that controls and streamlines the trading exchange itself.

The Tampa, Fla.-based company doesn't mind taking a behind-the-scenes role, and hardly a sexy one at that, as long as it helps generate a critical mass of customers who themselves create digital marketplaces and exchanges. Since mid-1995, it has been selling key ecommerce tools that can power many of the new B-to-B enterprises, ranging from restaurant supplies marketplaces to metal trading hubs. And the economic potential is huge. Tradex would be happy to settle with a fraction of the burgeoning business-to-business Internet commerce market, which Forrester Research estimates will mushroom to more than $1.3 trillion by 2003 (see p109 for more market data).

"[Tradex] would be sipping off the firehose of transactions taking place in the market, and that's not bad at all," says Bob Chatham, an analyst at Forrester in Cambridge, Mass. "It's definitely a viable niche they're in.... The question is how are they going to get there?"

Another enthusiast is Albert Pang, an analyst at International Data Corp., a research firm based in Framingham, Mass. "I'm very bullish on markets developing for these trading communities on the Internet," he says. "There are going to be many of these communities being formed using the different applications and services that Tradex can offer.... Once [companies] realize they can get savings out of buying and selling production goods, even feed for horse racing, Tradex will be in a good position to stake their claim to the market." He cautions, however, that the market for such electronic hubs is still small and immature.

Daniel S. Aegerter, Tradex's CEO, is applying his native Swiss penchant for efficiency and focus to the company, aiming to succeed in the nascent market by homing in on the exchange model and not trying to be all things to all players. "Our sustainable competitive advantage is that we're focused on marketplaces, on the external connections for buyers and sellers. I've preached this since 1995."

The company is targeting ebusiness portals and vertical marketplaces. The ebusiness portals are for existing businesses whose customers want to ".com" themselves and at the same time prevent their customer relationships from being eroded, Aegerter explains. Vertical markets are typically created by startups, companies without an existing customer base but with a deep level of expertise about trading among buyers and sellers. "These guys want to be the next Amazon," he adds. So what does Tradex want to be? "We want to be the engine behind leading marketplaces in every space," he beams.

In 1996, Tradex was spun off the U.S. subsidiary of Swiss company Dynabit, an importer and distributor of high-end computer peripherals for digital publishing. In 1995, the parent company introduced an Internet-based system to automate purchasing processes for its customers and suppliers. Aegerter, anticipating the huge growth potential in playing an intermediary role, initiated the creation of the marketplace and exchange technology that now serves as Tradex's core.

Tradex was ahead of its time in 1996, and in fact fumbled to find a foothold as the market evolved first from selling-related software, then to procurement software, before finally expanding to support out-of-the-box digital exchanges.

Digital exchanges are the more advanced — and hence unproven — B-to-B model, as outlined in "Let's Get Vertical" (p84). The first method to emerge for mediating transactions between buyers and sellers is the catalog model, which works best for small-ticket items and in markets where prices don't fluctuate too frequently. Companies such as Ariba, of Sunnyvale, Calif., and Commerce One, of Walnut Creek, Calif., dominate the space as vendors. Ariba's software automates the purchasing decisions of many Fortune 500 companies; Commerce One's software products automate supply-chain management of nonproduction goods and services.

Then came the auction model, which appears to work best for hard-to-find, used, or otherwise unsaleable products. One need only peek at the buzzing trading activity on eBay, Amazon.com, Onsale, and many others to see how popular this model already is in the consumer market. Tradex wants to dominate the third mechanism, the exchange model, by creating the digital engines that enable real-time matching of bid and ask processes for commodity goods. The stock and commodities exchanges are obvious precedents. Among those creating electronic exchange hubs are e-Steel in steel, and Tradex customer PlasticsNet.com in plastic materials.

"These guys have as good a chance as anyone for becoming the platform of choice for vertical market exchanges," says Tom Kippola, managing partner of The Chasm Group, a high-tech market strategy consulting firm in San Mateo, Calif. "The whole idea of vertical market exchanges is emerging but hot, and it'll get hotter. Companies are going to need a platform to run their business." Kippola has consulted for Tradex and is mulling an offer to join its board of directors.

Several companies already call upon Tradex to run their ecommerce hubs. The largest customer is Nippon Telephone & Telegraph, which has built an internal marketplace to aggregate its companywide expenditures for maintenance, repair, and operations (MRO), goods and related products. The solution allows NTT's myriad business units and

subsidiaries to create a central catalog of approved products, suppliers, and contracts. Other customers include PlasticsNet, MetalSite, and the ecommerce site Hospitality Online Trading System (HOTS ecommerce), an Australian food-service marketplace.

A more recent customer is VerticalNet, which builds communities for corporate buyers and sellers (see "Industrial-Strength Portal," p118). The company is piloting Tradex's technology for a few months to help manage 1,000 customers and potential customers, who together run thousands of storefronts. C.H. Low, chief technology officer of VerticalNet, says Tradex's software lets it manage the B-to-B partner relationships in a scalable fashion, and its dynamic user interfaced (a JavaBeans-based product) makes it easy for customers to customize their own processes.

Tradex product Commerce Center 6.0 helps customers handle registration for large marketplaces and manage contracts, pricing, and work flow. Forrester's Chatham gives it a thumbs up. "Their architecture is superb. It gives the right level of customization and it's a great platform to build on, with a strong set of components," he says.

To date, Tradex makes money from licensing its software (60 percent of revenues), from support services (30 percent) and from charging transaction fees (10 percent), according to Aegerter. He expects the company will rely increasingly on transaction fees for income down the road. That will give Tradex a larger interest in customers' luring more participants to their trading communities. "We'll grow on a revenue-sharing basis far more than on up-front fees," he says. Aegerter is coy when discussing the privately-held company's financials. In its June quarter Tradex boasted a 70 percent sequential hike. He would not project when the company's bottom line will shift from red to black. "Wait for our S-1," he says.

Its piggy bank of $20 million — from venture firms Draper Fisher Jurvetson, Internet Capital Group, and Sigma Partners — should keep it afloat for months even if the window for Internet initial public offerings closes.

If the recent dizzying IPOs by Ariba and Commerce One are a reliable barometer, Tradex could ride their coattails. "Ariba's IPO is great for us," Aegerter says. "It establishes a great comparable." Then again, Tradex could be at a disadvantage since Ariba and Commerce One beat them to the punch, as Mohanbir Sawhney believes ("Let's Get Vertical," p84). Ariba's stock was priced at $23 on June 22, started trading at $61 the following day, and had doubled that by mid-July. Commerce One's debut has been a bit more modest; it was priced at $21, started trading on July 1 at $66 and was idling right around there in mid-July.

Challenges loomAegerter acknowledges he's got a long way to establish Tradex as a high-profile, successful player. Toward that end, he says the company must partner with other companies to provide more functions for vertical hubs, including auction capabilities, despite the company's commitment to focus most on the digital exchange space.

Meanwhile, larger players in the general B-to-B marketplace-enabler arena, including Ariba, Commerce One, and to some degree the smaller EC Cubed, are expanding their

offerings, making them both potential partners and rivals to Tradex. "Tradex will also want to expand its offerings, and to that end it's likely to look for partners or possibly even acquisitions," predicts Sawhney.

C.J. Glynn, manager of corporate marketing at Ariba, is zip-lipped about Ariba's acquisition plans, but he did say "we're open to partnering with people who could provide pivotal end-to-end ecommerce solutions." As for Ariba's interest in expanding into the exchange model, "I wouldn't rule it out for Ariba," he says.

Lauren Whitehouse, vice president of marketing at EC Cubed of Wilton, Conn., refers to Tradex as "somewhat of a competitor. Though we have a very different solution, we compete for the same IT dollars." EC Cubed's product, ecWorks, is a suite of Java components designed to identify common functions required for purchasing goods, digital marketplaces, supply chain management, and customer service. Aegerter appears unfazed by such potential rivals. "The ones on our radar really are people who decide to build [platforms] by themselves," he says.

So continues his mission to convince high-volume brokers of commodities ranging from cow feed to cowhide to derivatives that their future lies in creating vertical marketplaces on the Internet.

VerticalNet: Industrial-Strength Portal From paint to power plants, waste to widgets.

By Eric Hellweg

Hold on a second-I'm trying to find my car," Mark Walsh, CEO of VerticalNet, barks into his mobile phone, pacing around the parking lot of his Horsham, Penn. headquarters. "OK," he finally says, "I think I found it." After a moment's laughter, he confesses, "I always thought I'd have a CEO's spot right in front of the building like they used to have at banks, but I'm finding the Internet has zero patience with that kind of perk bullshit."

Parking lot privileges aside, the Internet does deliver other perks with an alacrity well suited to those with little or no patience. Barely four years old, VerticalNet, a creator of online vertical trading communities, has been rewarded with some of the ultimate Net perks: a market cap nearing $2 billion and an IPO that stands among 1999's most successful. Its stock skyrocketed from an offer price of $16 a share in early February to a high of $140 in just over two months. Not bad for a company whose CEO proudly proclaims it "the least sexy site on the Internet."

In an age when many Internet properties are valued largely on their flash or the loftiness of their business model, VerticalNet makes money paving new sales avenues and creating communities for 41 unglamorous industries, such as food ingredients, paints and coatings, and adhesives.

"We're doing what the Net is supposed to do-expand your buying community and drive transactions," Walsh says. In doing so, VerticalNet has become a model for business-to-business infomediary success. Each of VerticalNet's arcane-topic portals serves as a hub for the denizens of each trade. People interested in solid-waste management can visit www.solidwasteonline.com and read pithy news stories, product reviews and recommendations, and visit storefronts from hundreds of vendors.

"VerticalNet is in many ways a poster child for infomediary success. They are essentially a portfolio of over 40 vertically focused infomediaries. It's a mutual fund of infomediaries," says Charlie Finnie, managing director and senior research analyst at Volpe Brown Whelan (see "Infomediaries on the Move," p123). "The company is bringing entire industries online that heretofore couldn't even spell the word Internet."

Buyers at VerticalNet portals know that all the vendors presented are in their field of interest, and by virtue of the store's placement, nearly every lead the seller gets is a qualified one.

These thousands of storefronts account for 65 percent of VerticalNet's revenues. About a quarter of revenues come from sponsorships, and about 10 percent come from commissions on transactions enabled by VerticalNet.

The company charges $6,000 per year to host the storefront, "a figure that's trending upward nicely," Walsh notes, and the response has been positive. "We have a 90 percent

renewal rate with our storefronts. Our salespeople can barely keep up with the vendors begging them to put a 'buy this now' button next to their products." The storefronts have also proven successful as low-cost sales methods, allowing companies to reach buyers in geographic areas or industries in which they have no sales force.

A VerticalNet-sponsored survey following up on people who left sales leads in storefronts found that 15 percent bought the product from the manufacturer, and 25 percent said the product was still under significant consideration. Fifty percent of the time, these purchases were made from a company the buyer had never done business with before. The average dollar volume of the transaction? $25,000.

One would think that VerticalNet would ask for a cut of all of these transactions it's enabling, but that's not the case. According to Walsh, the company doesn't want to risk damaging the relationship with its clients this early in the game. So instead, it has gone the M&A route to offer new services. For example, VerticalNet bought Isadra, a Palo Alto, Calif.-based industrial catalog company. VerticalNet is also investing roughly $1 million in Tradex Technologies (see "The Enabler," p112), whose procurement software platform VerticalNet is already licensing. "As we deploy these things, we expect a portion of the sales," says Walsh.

In 1998, the company lost $13.59 million on revenues of $3.13 million, and reported a 37-cent loss per share in 1Q 99. "Analysts [predict] that we should be profitable by the end of 2000 or early 2001," Walsh claims. "Some of our individual portals are profitable, though."

No www.lemonade-stands.comBefore any industry becomes a portal on VerticalNet, it must meet certain key requirements. There need to be at least 3,000 companies in the industry, and those companies need to have 40,000 buyers. Thus insuring "a fragmented market with both buyers and sellers that need to find each other," explains Mike McNulty, senior vice president and co-founder of VerticalNet. The industry as a whole also must have $10 billion in domestic sales, and a $10 million trade publication ad budget. The segment must also have a high penetration rate of Internet-ready computers.

So far, VerticalNet has found 41 such industries that meet these requirements, with rollout plans for the rest of the year. "We hope to have 50 by year's end," Walsh says.

Industry watchers are cautiously optimistic about VerticalNet's prospects. "I'd be surprised if every single vertical turned out to be a home run," says Finnie. "If even half of the verticals turn out to be leaders, VerticalNet will be one hell of a company."

As is the case with nearly every community-based infomediary, stickiness is a top priority. VerticalNet has introduced auctions into its sites, with moderate success.

"We're still extremely pumped about the auctions," Walsh says, but admits that "getting registrants and bids is a little tough. We had some problems with our user interface, so we're fixing that. We're about to announce a deal with a used-processing equipment auction company that will be similar to the eBay/Butterfield deal. They're going to give

us hundreds of millions of products on their shelves and [we'll give them] guaranteed bidding through us."

Don't expect to see Furbies or Beanie Babies on a VerticalNet auction, though. The company shoots for higher-ticket items. At the end of the summer, for example, poweronline.com will auction off three, $38 million power plants.

A vertical worldJust as VerticalNet-through its various portals-is providing companies a low-cost way to reach customers they've never done business with before, the Internet itself is providing the same service to VerticalNet. "With absolutely no promotion, we've generated 40 percent of our traffic outside the United States," says Walsh.

At the end of June, the company announced its first international partnership, with South Africa's Metropolis, which will pay VerticalNet an upfront fee to roll out South Africa-focused versions of VerticalNet's portals.

VerticalNet faces several challenges, however. Business-to-consumer sites are able to wrangle consumers away from their previous buying places a great deal easier than B-to-B sites. Consumers usually need only get approval from themselves to buy something. Business purchasers seek approval from chains of command before making a purchase. "Legacy behavior is far more difficult to change than legacy systems," Walsh admits. Volpe's Finnie also thinks that VerticalNet needs to build depth as well as breadth.

"VerticalNet is not going to attack every market," says Finnie. "If there's strong competition in a certain market, they may say it's not worth the trouble. If a company like e-Steel gets traction and is the dominant place to go, I think VerticalNet will think twice.

Finnie is nevertheless bullish on the prospects of B-to-B infomediaries. "Most major industries will have a dedicated infomediary in three years," he says. "Five years from now, those infomediaries will have played a role in the majority of transactions."

To Walsh, this and other pressures for VerticalNet's success are somewhat inherent. "God knows, our stock symbols is VERT, which is French for 'green.'"

B2B Commerce: The Next Frontier

Infomediaries on the Move EXCLUSIVE: Highlights from a new B-to-B commerce report by Volpe Brown Whelan

How Big is Big?

B-to-B infomediaries employ a variety of mechanisms to generate transaction fees, including: auctions; matching buyers and sellers through request for proposals/quotation services; and direct sales of merchandise such as books and software. Infomediary transaction fees will grow as businesses conduct more of their commerce on the Internet, and as infomediaries account for an increasing share of this commerce.

Total business-to-business Internet commerce — defined as the value of all goods and services purchased over the Internet by business users (excluding advertising revenues) — has grown from virtually zero in 1990 to an estimated $43 billion in 1998, and is projected to grow to $843 billion by 2000, according to Forrester Research. We believe that infomediaries will account for more than a quarter of all B-to-B Internet commerce revenues by the year 2002, and estimate that the gross value of transactions completed by B-to-B infomediaries will grow from $750 million in 1998 to approximately $211 billion in 2002.

Early evidence indicates that infomediaries will be able to charge transaction fees ranging between 5 percent and 10 percent of the dollar amount of transactions they complete. Taking the low end of the range (5 percent), the infomediary transaction market will grow from $38 million in 1998 to $10.5 billion in 2002. Investors should look at the transaction fees, not the gross value of transactions, when they examine infomediary transaction-based revenues, since transaction fees tend to carry a very high gross margin, typically 85 percent or higher, as eBay shareholders are well aware.

Why Buy? Business-to-business infomediaries have the opportunity to benefit from the same increasing returns dynamic as the fax machine did over the last decade. Three factors also make them compelling to investors: a highly profitable business model; rising barriers to entry; and huge market opportunity.

Once an infomediary achieves its critical mass of networked users, as the fax did in the late 1980s, there is every reason to believe that the infomediary's revenue will accelerate, as the following cycle plays out:

New buyers attract more new buyers and vendorsJust as fax machine sales accelerated when critical mass was achieved, infomediaries will enjoy an acceleration in membership once end-users decide that a particular infomediary is the place to go in a specific vertical market. Once the eyeballs begin to congregate at a particular infomediary, vendors will surely follow.

Buyers spend moreBuyers will concentrate their spending where they find what they want and need. The infomediaries that take an early lead in attracting buyers and sellers will capture an increasing share of transactions and advertising revenue. Further, buyers, at the margin, will spend more of their budget as two things happen: buyers gain awareness of new products through the wider reach and cheaper distribution provided by the Internet; and buyers repurpose the money saved through reduced transaction costs.

Who Wins?

Those in large and fragmented marketsFragmented markets provide the competitive dynamics that favor the development of info-mediaries. Markets that are not fragmented are instead controlled by either a buyer or a seller (or a group of buyers or sellers), and are less likely to foster the development of infomediaries. Infomediaries create the most value in markets where buyers and sellers have difficulty finding each other.

Those with sharpest focusInfomediaries should focus on solving a particular problem for a particular vertical market. By declaring a distinct focus area (e.g., laboratory supplies), the infomediary attracts buyers and sellers whose primary interest is that particular subject area. But if the infomediary's category is too broad or murky (e.g., "supplies" instead of "laboratory supplies"), almost no one (buyers or sellers) will feel a sense of urgency to go there. By sharpening their focus, infomediaries can provide a depth of information — and hence customer loyalty and participation — about a particular subject that is unrivaled. Increased customer participation adds yet more depth to a firm's knowledge base, which in turn drives more transactions.

B-to-B Bull: Charlie Finnie

Occasional flashes of prescience mark the 13-year record of San Francisco- based investment bank Volpe Brown Whelan — most prominent being its underwriting of Netcom Online Communications' initial public offering back in 1994, the first Net-stock IPO.

And as one of many tech-focused investment firms that thrive in Wall Street's shadow, it has to pick its targets — typically technology and health care — carefully. All of which might make you wonder — at first — why the firm is putting a lot of money where managing director Charlie Finnie's mouth is.

Finnie, 40, has emerged as one of the country's most forceful B-to-B ecommerce proselytizers over the last couple of years, boasting frequently in the press that this emerging generation of B-to-B companies — including Volpe-backed firms such as VerticalNet (see "The Industrial-Strength Portal," p118); online commercial real-estate matchmaker Comps.com; and RoweCom, a digital subscription house — will "be the driving force behind the economy for the next decade."

In an expansive new Volpe report on B-to-B ecommerce to be released this fall — previewed here exclusively for Business 2.0 readers — Finnie claims that the new online army of B-to-B middlemen will become "the fastest-growing and most profitable business model in the Internet arena over the next 3 to 5 years."

Big words, those, coming from a guy whose macroeconomic wisdom began as a 27-year-old Booz-Allen & Hamilton research analyst in 1986, helping out on a mammoth study for a major fast-food chain that tackled a vexing question: What color wrapper to sell fish sandwiches in? (Blue.) Next, as a securities analyst with Alex. Brown & Sons in 1987, Finnie admits he still "barely knew what a balance sheet was."

A decade later, Finnie has had time to do his homework. Today, he helps guide Volpe's Net investment strategy in the burgeoning business-to-business realm. The reasons for his speculative zeal? The market opportunities awaiting in the B-to-B Internet space, he says — from manufacturing to utilities — are both higher-impact and higher-margin than their consumer-market counterparts. From a revenue standpoint, he explains, B-to-B infomediaries "don't have to attract a billion eyeballs that are going to spend 10 bucks on you." In multibillion-dollar industries where average transactions can run more than $10,000 and an infomediary's profit margins are a lot fatter than those of books or CDs, "you can attract a smaller number of eyeballs, all of which are attached to much deeper pockets." Along with promising projections for B-to-B infomediaries — a $20 billion chunk of all B-to-B ecommerce by 2002 — the report deconstructs some of the unique investment metrics in this new breed of companies. — Jeffrey Davis

First moversIncreasing-returns business models favor the company that moves first, and moves decisively, in a particular market. Companies with more customers will attract a disproportionate share of customers; infomediaries exaggerate this effect, as their cost of customer acquisition becomes incrementally lower over time than that of an offline business. Further, infomediaries tend to develop high switching costs as a natural part of their evolution. Both of these points back up the argument that it is better for investors to pick the first-mover infomediary in a particular vertical market.

Big capital spendersInfomediaries need capital to acquire a customer base; they have to pay more to acquire members than offline counterparts. In the early stages of developing a vertical market, competition for customers may increase, and marginal cost of each new customer may rise. Eventually, a market leader's customer base will reach critical mass and the marginal cost of acquisition will decline. Winning this battle requires substantial up-front spending on brand building and customer acquisition.

Developing the basic technology required to operate an infomediary is an expensive undertaking. The systems required to create open exchanges, updated product catalogs, community features, data collection, and analysis tools all take time, money, and effort to build. And the integration step, where an infomediary builds its systems into those of its customers, can be very expensive.

Switching costs riseIn the world of infomediaries, where operating costs are low compared even to increasing returns offline businesses, the decrease in operating costs is magnified. The primary cost for infomediaries is acquiring buyers and sellers. Initially, the infomediary loses money on every customer acquired as customer revenue is far less than customer acquisition cost, resulting in a net loss for the infomediary. But each new customer drives revenue per customer upward. The result is that the eventual profit dramatically eclipses earlier losses.

Creating switching costs for B-to-B infomediaries is different than it is for B-to-C infomediaries. Luckily for the B-to-B firms, switching costs are less subtle and probably more permanent than those in B-to-C. But they are also more difficult to build.

B-to-B infomediaries build switching costs by integrating themselves into the systems and/or processes of the buyers and sellers that they serve.