Is there an Amazon.com for Every Industry?
By Stewart Alsop

(FORTUNE Magazine) – Are there any industries that should not be Amazoned? This is a big question for a venture capitalist. As you well know, electronic commerce seems all the rage these days. For the most part, it means finding a way to sell stuff over the Internet. Amazon.com figured out how to sell books on the Internet--it figured that out really well. So now venture capitalists are getting business plans for new businesses that want to sell everything else on the Net. Stocks are taken (E*trade, Schwab, etc.). So are music CDs (CDNow, Music Boulevard, and Amazon.com) and videos (Reel.com and good old Amazon). Those are the easy categories. The new businesses aim to apply the same principles to homes, home mortgages, furniture, consumer electronics, insurance, and so on--stuff that doesn't easily go into a small box for overnight delivery by UPS or Federal Express.

These more recent business plans say something like "Amazon.com had it easy, because the book business is so highly concentrated. We're going to be more successful than Amazon because the industry we have targeted is highly fragmented, and that makes it ripe for disintermediation." These entrepreneurs must think venture capitalists are impressed by big words, and judging by the amount of money being invested in these businesses, they might indeed be right! But that's all they are right about.

The fact is, their new favorite big word--disintermediation-- may be seriously overrated.

I've got this theory aborning: the opportunity for disintermediation exists in direct proportion to the degree of concentration in an industry. In other words, the arrival of the World Wide Web and electronic retailing is not a harbinger of imminent concentration in fragmented industries. Instead I'm beginning to believe that there is a reasonable chance that the Web will enhance the role of middlemen in complex markets, just as it reduces the role of middlemen in concentrated markets.

Just to review, remember that disintermediation is a highfalutin word for the process that leads to the elimination of intermediaries in an economic system. Say you start with a system that calls for the people who make something to sell that something to people who ship it, who then sell it to people who resell it, who then sell it to people who actually consume it. The height of disintermediation would be a change that led to the people who make something selling it directly to people who consume it.

Disintermediation has positive effects: lower costs of distribution and happier customers who get exactly what they want for less money and in less time. It also has negative effects, including serious dislocation as intermediary companies disappear.

The World Wide Web is the embodiment of disintermediation. With the arrival of a truly ubiquitous, universally accessible network, the theory goes, companies can provide consumers with precisely what they want at the lowest possible cost with the shortest time between manufacture and delivery. With that kind of technology, industries will become concentrated and efficient.

The problem is that the industries that appear to respond most readily to this are the ones that were streamlined before the arrival of the World Wide Web. Technology played a big role in that concentration, but the opportunity to revolutionize industries was so great that companies spent the money to build proprietary infrastructures for competitive advantage: Barnes & Noble in the book industry; Home Depot in the home hardware industry; Wal-Mart in the household goods industry; Circuit City in the consumer electronics business; and so forth. These companies, along with their biggest competitors, decimated the networks of small retailers and large distributors in their industries.

Then Amazon.com came along. Here's a company that doesn't have to build stores or train retail salespeople or build a distribution network or store as many books. Not only that, but it can be a whole lot more efficient in helping customers get exactly what they want. Hooray! Goodbye, Barnes & Noble and Borders! A lot of people mistake this Amazon.com phenomenon for an example of disintermediation caused by the arrival of the World Wide Web. But the truth is that Amazon.com is just another superstore--albeit one that doesn't have to invest in building stores, managing inventory in those stores, or staffing them.

I keep thinking, "Why books and not some other business?" Yeah, maybe music and movies fit, although that's still up in the air. Stocks seem to fit pretty well too. And flowers.

But what about the stuff that's hard to buy, where the technology would seem to create real savings: home hardware, household goods, furniture, consumer electronics, drugs, toys, home mortgages, automobiles, insurance, health care? All are areas in which entrepreneurs are using that "disintermediation" word to separate venture capitalists from their money, and in which one or more companies are taking a shot at being the next Amazon.com.

Our firm recently evaluated an investment in a company that makes a different proposition in the home-mortgage business. It would compete against companies like E-loan and Homeshark, which provide consumers with the opportunity to compare rates from different lenders and secure a loan online. I won't name the company, since we're discussing a possible deal. But the premise is that mortgage brokers are a crucial piece of the home mortgage system, because mortgages are so complex and the system for granting mortgages so byzantine that customers need a knowledgeable adviser. So this company has developed software specifically designed to help the mortgage broker, not the mortgage consumer. It argues that E-Loan's and Homeshark's model of providing a direct interface between consumer and lender only works for refinancing existing homes, when the consumer doesn't have to split attention between buying the home and financing it.

When I heard that pitch, it resonated. We used a mortgage broker when we bought a house a year ago. I remember feeling so completely overwhelmed with the details of the purchase that I couldn't contemplate also seeking out a mortgage simultaneously.

The combination of that experience and the mortgage entrepreneur's pitch got me thinking. I've seen pitches for companies that want to be the Amazon.com of virtually every category I named above. Maybe that's not a wise model for complex industries. Maybe the only way Web retailing really works is when an industry has already been disintermediated and is highly concentrated. Maybe the brilliance of Amazon.com was that it didn't try to use technology to redefine an industry, but to take advantage of an industry that was already headed in that direction. Indeed, it may be that venture capitalists should stop trying to find the next Amazon.com and instead look for ways to infect complex industries with new kinds of software--the kind of software that helps middlemen instead of trying to eliminate them.

STEWART ALSOP is a partner with New Enterprise Associates, a venture capital firm. Except as noted, neither he nor his partnership has a financial interest in the companies mentioned. Alsop may be reached at stewart_alsop@fortunemail.com; the column may be bookmarked at www.fortune.com/technology/alsop/.