(FORTUNE Magazine) – If you asked Internet retailing experts three years ago who was most likely to suffer from the rise of e-commerce, many would have pointed to distributors. These large disrespected behemoths that move physical inventory were presumably breathing their last breath. As e-commerce vendors gained steam, they would order products directly from manufacturers--and the distributor would be rendered unnecessary. The interesting fact is that instead of disappearing, some distributors are thriving. Though the tectonic plates of online retailing are still shifting, the end result may look far different from how it does today.

Before looking forward, let's examine the world of e-commerce. Consumers shop at online stores. Many times, they are directed to these stores by portals, such as Yahoo, or affiliates--small sites on the Web (over one million of them) that advertise products and get a commission for sending leads to e-tailers. When consumers purchase products, many e-tailers fulfill these requests through distributors. These distributors perform as they always have, securing products from the end manufacturers.

The interesting thing about this value chain--manufacturer, distributor, e-tailer, portal/affiliate, consumer--is that it hardly resembles the lean economic model promised in the early days of the Internet. With five parties and four transfer points, there are many mouths to feed. So how will this model evolve? Which points are likely to be eliminated? To many people's surprise, with the emerging importance of portals and affiliates, distributors could remain strong. In certain cases a likely supply chain model may be manufacturer, distributor, affiliate, customer--where the e-tailer is the one eliminated. I suspect this sounds a bit heretical in Silicon Valley circles, but give me a chance to explain.

Ingram Micro, in Orange County, Calif., is the largest distributor of computer hardware and software on the planet. Traditionally, Ingram has aggregated products from manufacturers and delivered them to storefronts, which would then resell the product to the customer. However, in response to the rise of e-commerce, Ingram has changed its colors. Rather than focus simply on bulk distribution, this savvy distributor can now ship one unit directly to the customer. What's more, it can make it look like the order came from someone else, complete with customized packing slip.

If this weren't enough, Ingram has also created an extranet that shares merchandising information--product descriptions, photos, spec sheets--with anyone who wants to create a virtual Web storefront for computer products. The really interesting thing here is that as Ingram hones its model, it increases its position on the value chain. There is no reason its affiliates or sites like Yahoo and AOL can't simply send orders directly to it as opposed to routing them through an e-tailer.

Ingram is not the only Web savvy distributor. Distributors in the book, music, toy, and office-supply industries are building expertise in "unit of one" shipping. As these skills continue to improve, there could be some fundamental shifts in the overall e-commerce environment.

The first shift to expect is a bifurcation in the e-tailing community. The industry will need to differentiate between e-tailers who handle inventory and physical distribution (physical e-tailers) and those that don't (virtual e-tailers). Many people believed that virtual e-tailing was the optimal way to go, because of the lack of investment in physical assets (both products and warehouse infrastructure). However, because of the evolution outlined above, the virtual e-tailer may be in the most tenuous position. The barriers to entry in order taking are simply too low.

Because virtual e-tailers don't own their inventory, they are likely to be faced with several unfortunate realties. Their value added is really limited to demand aggregation and order taking. So as distributors increase power, putting an order through a virtual e-tailer will be of marginal value. As a result, the economics of a virtual e-retailer should begin to mimic those of a portal or an affiliate. The gross margin of an e-tailer and the commissions paid to an affiliate (typically 5% to 10%) will undoubtedly converge (if they haven't already). Moreover, virtual e-tailers are likely to find more competition as their distribution partners enable others to take orders in an online world.

Physical e-tailers, such as Amazon.com, are in a much more defensible position. This is particularly true in markets where traditional distributors are fragmented, behind the times, or nonexistent. Good examples of these types of industries include jewelry, luggage, and sporting equipment. In these markets physical e-tailers are typically the first players to aggregate products at a single point, and by building for unit-of-one distribution from the beginning, there is a chance for better optimization in the warehouse. In addition, by controlling both the ordering and fulfillment processes, there is a major opportunity to excel at customer service vis-a-vis the distributor-based model. Over time, physical e-tailers will likely find themselves competing with distributors, which is ironic, as many of them were originally enabled through partnerships with distributors.

My firm has recently invested in Vstore, a company in Stamford, Conn., that aims to connect these new-age distributors with individuals, affiliates, and portals. Through Vstore, anyone that wants to engage in e-commerce order taking can do so. Vstore has three primary assets. First, it has a robust back end, complete with links that efficiently route orders to the appropriate distributor. Second, it has a merchandising engine that provides product information, photos, and technical specifications to the affiliate or order taker. Last, Vstore has created an amazingly simple tool that lets anyone quickly build a professional storefront--products, ordering capability, and customer service included.

With Vstore acting as an e-commerce infrastructure provider, content sites and affiliates can create their own themed storefronts with products that cut across many categories. For instance, if someone wanted to create a Web store dedicated to Elvis, he could use Vstore to aggregate books about Elvis, Elvis movies, Elvis figurines, Elvis T-shirts, and even Elvis toys. He could also include relevant items such as books about Tupelo, Miss., Elvis' birthplace. Prior to the emergence of such networks, these affiliates were forced to send orders to multiple e-tailers that each specialized in a particular product category. This resulted in a plethora of brands on the affiliate's Website and, even worse, forced the affiliate to send customers away from its own site. By eliminating the virtual e-tailer in the economic chain and taking orders directly to the savvy distributor, we have achieved a more efficient supply path.

As electronic order taking becomes easier, there will be increased value placed on those businesses that excel at aggregating orders, distributing physical products, and facilitating the customer purchase decision. This will include companies with large amounts of traffic, such as portals like Yahoo and AOL, as well as new content companies, like eHow, epinions.com, and Respond.com (my firm has investments in the latter two), that provide research and how-to advice for those purchasing products. Those who just concentrate on building an e-commerce storefront may not be so lucky. Taking an order and putting it in a shopping cart is already a commodity.

J. WILLIAM GURLEY is a partner with Benchmark Capital, a venture capital firm. Except as noted, neither he nor Benchmark has a financial interest in the companies mentioned. To receive an expanded version of Above the Crowd, visit www.news.com, or to subscribe to the e-mail distribution list, please send an e-mail to subscribe-above_the_crowd@atc.revnet.com. If you have feedback, please send it to atc@benchmark.com.