*[A version of this article originally appeared at Knowledge@Wharton on May 11, 2011]

Analysis on the increasingly lucrative field of internet commerce.

A decade after pets.com and a string of other early Internet specialty retailers collapsed, a new wave of start-ups — enabled by the power of cloud computing, advanced delivery systems and deep social relationships with customers — is shaping e-commerce.

From diapers and eyeglasses to pool tables and, yes, pet products, entrepreneurs are developing specialty businesses to compete alongside one-stop shopping giants like Amazon.com and Walmart.com.

"There is a new generation — Internet retail 2.0," says Wharton marketing professor David Bell, noting that after the 2000 dot-com bust, activity in online specialty retailers dried up as financing became difficult, or impossible, to get and industry executives struggled to evaluate failed business models. Now, engineers have dramatically reduced obstacles to creating web-based businesses, and entrepreneurs have learned more about how to capture online consumers.

Many of these small players are working in affiliation with Amazon or other major online retailers that assist with the marketing platform and/or delivery fulfillment. Others are folding into bigger sites. This month, Seattle-based Amazon paid $545 million to acquire Quidsi, a New Jersey retailer that delivers diapers and baby products under the name diapers.com and recently started selling soap and cosmetics. Last fall, Amazon paid $1.2 billion for Zappos, the specialty shoe retailer.

Online sales continue to outpace overall retail spending growth. In 2010, U.S. e-commerce sales totaled $165.4 billion, up 14.8% from 2009, the U.S. Department of Commerce reported. Looking forward, eMarketer, which tracks the digital marketplace, estimates online sales will rise to $188 billion in 2011 and $269.8 billion in 2015. According to eMarketer, 87.5% of U.S. Internet users over age 14, or 178.5 million people, will browse or research products online this year. Of that group, 83% will make an Internet purchase in 2011.

The Internet lends itself to specialty retailing because it allows companies to offer huge product portfolios without having to stock inventory in bricks-and-mortar shops. Wharton marketing professor Stephen Hoch notes that Amazon itself began in 1994 as a niche player selling books only. Now, the $34 billion company markets everything from fine jewelry to industrial supplies.

According to Hoch, improvements in computer technology and graphics allow consumers to get a better sense of what they are buying online. "People, especially young people, have grown up on this and are more trusting and more confident about buying online." Zappos, for example, made it easy for consumers to buy shoes — "the one thing that really has to fit right" — online with liberal return policies. And eyewear e-tailer Warby Parker will ship sample frames to customers. The site also has an interactive feature that allows shoppers to upload their photo to see what they would look like in different styles.

In the earlier days of online retailing, Bell notes, companies overestimated the value of "eyeballs," or the number of visits consumers made to a site. Retailers now understand more about converting visits into sales, often by building relationships through social commerce tools, including blogs, Twitter and Facebook. "Companies are learning how to turn their own customers into their sales force," says Bell. "The whole social media platform wasn't around in the first boom."

Jeffrey Grau, principal analyst at eMarketer in New York City, predicts that social connections will continue to drive growth for Internet retailers as mobile technology expands. "There are a lot of creative business models coming out all the time that involve social commerce or mobile commerce. There's a lot of innovation going on in e-commerce."

Pitching to New Parents

But Wharton marketing professor Peter Fader cautions that e-tailers hoping to evolve small niche lines into national mega-brands face difficult, if not impossible, odds. He cites the marketing law of "double jeopardy" and research showing that big brands enjoy higher sales and more brand loyalty than their smaller, more specialized counterparts, which tend to appeal only to the most avid buyers.

He uses the example of a consumer who buys a specialty brand of gentle laundry detergent for people with sensitive skin. That person is probably still buying name brands such as Tide to meet his or her other needs. A business based on the specialty detergent could never scale up enough to capture significant sales from the major brands, Fader says. Online specialty retailers can be successful if they understand their limitations, and construct logistics and marketing operations to suit that size.

"Unfortunately, too many small brands don't view themselves as a specialty," Fader points out. "They think they can compete with the big guys." As a result, the company spends too much on broad-based advertising and other attempts to drive scale. Ultimately, Fader says, they "get crushed…. If you admit that you are a small brand, go after the specialty angle and say, 'People will buy us only occasionally' — and you keep costs down and keep the message focused — then you can be okay." However, he adds, "That [approach is] not sexy and doesn't sell well to the venture capitalists."

According to Wharton marketing professor Leonard Lodish, diapers.com retailer Quidsi, founded in 2005 by Marc Lore and his business partner Vinit Bharara, has the important characteristics for an online specialty retailer today — strong execution and lavish customer service. Lodish, who served on the board of Quidsi before its acquisition by Amazon, says the firm developed highly efficient distribution networks revolving around state-of-the art information technology. Its warehouses are equipped with sophisticated robots handling automated fulfillment. In addition, Quidsi developed a software program, called Boxem, that calculates the best sized box for any individual order, saving on shipping and other costs.

However, Lodish says the company's most important asset is its relationship with new parents, fostered through its own customer service staff, not a remote call center. "Their customers love them," notes Lodish. "Their reputation with young mothers is better than anybody else by far. People talk about them and refer friends to diapers.com when they have new babies."

After battling Quidsi on diaper pricing, Amazon offered to buy it. "The present value of what Amazon was offering was better than what [Quidsi] could do on its own — That's the right way to look at it from the point of view of the shareholder of the company," says Lodish.

Wharton marketing professor Barbara Kahn describes Amazon's purchase of Quidsi as the online version of a long-established retail phenomenon known as the "wheel of retailing" in which large companies seeking scale and new capabilities often acquire specialty retailers to expand their offerings in a one-stop environment.

However, in time, new specialty retailers discover niches and create targeted organizations to meet those customer needs better than the large retailers — and the wheel turns again. The differences between a company like Amazon and a small specialty player illustrate the disparities between a market-share model, designed to serve as many products to as many people as possible, and a customer-focused approach aimed at developing deep loyalty from shoppers, according to Kahn.

The new breed of Internet specialty retailer is reaching customers by building communities among customers, she adds. In traditional retail, the closest example of this kind of effort is at independent running stores where customers come together to form running clubs and train for races. "In the virtual world you can do that around any interest," Kahn notes.

Three Engineers vs. 40

Social commerce is critical to Davis Smith's latest venture. In 2004, he founded pooltables.com, which shipped pool tables made in China to consumers' homes. Smith sold that business to a competitor to focus on his next company, an online baby product retailer in Brazil called Baby.com.br — modeled largely on diapers.com. While the margins on diapers are slim, Smith says, the new company's strategy will be to develop loyalty so that mothers return to the site frequently, not only for diapers but also for wipes and other baby products.

"In these niche markets, customers, especially mothers, expect great customer service and [significant] depth of products and they actually want a community," Smith says. The site, scheduled to open this spring, will have outlets for consumers to share their pregnancy and parenting experiences. "This whole idea of social commerce is very new. No one has really figured it out entirely yet. But a lot of people are coming in brand new with a lot of fun ideas."

Once he has a customer hooked, Smith figures families with two children will be loyal customers for at least six years. "When you find mothers coming back to the website to buy the wipes and the formula they need every week, there is a tremendous opportunity to up-sell and cross-sell other products." Those same mothers, he adds, could be converted to other consumer products because "they are not going to want to strap their toddlers into a car seat to run out to buy q-tips or Windex or toilet paper."

Beyond logistics and customer relationships, some of the new web entrepreneurs are focused on creating their own products. Bonobos, a men's apparel e-tailer, designs, develops, sources and merchandises its own clothing brand. "So unlike Zappos or Quidsi, we are not about offering the deepest selection of brands on the market," states Bonobos founder Andy Dunn, who started the company in 2007. "Rather, we want to curate brands, and go so far as to actually create and own the very best of those brands." Dunn calls the strategy "vertical web retailing."

Even pet food is getting a second look. Petflow.com, founded in 2009, is delivering food and other pet supplies on a new subscription model that relies on regularly scheduled deliveries. Julie Wainwright, former CEO of Pets.com, says specialty retail start-ups today are operating in a completely different world than when Pets.com folded after famously promoting its business with commercials starring a sock puppet during the Super Bowl.

First, Wainwright notes, the number of people with Internet access, including mobile devices, has grown from 250 million in the early part of the decade to around five billion today. In addition, technology has evolved with new plug-and-play solutions that had to be custom designed for e-commerce firms in the first wave of online retail. For example, new start-ups today can buy a digital shopping cart from a software company. Pets.com had to design its own. Wainwright says three people can now do the work of creating a site that took 40 to 45 highly skilled engineers a decade ago. Back then, cloud computing was unknown. Each individual e-commerce site had to operate its own servers and backup systems.   "Now, it is pretty easy for any mom and pop to set up an e-commerce site," states Wainwright, who is currently running a health consultancy.

Wharton's Bell says that in addition to mom and pop entrepreneurs, many top business school graduates are looking for ways to start their own firms rather than migrate to finance as they had in recent years. And as Wainwright notes, online retailing today is becoming an increasingly creative outlet. Rather than simply moving products out of warehouses, online retailers now have the opportunity to build elaborate sites incorporating photos, video and music that can express a personal vision.

Subsidiary or Competitor?

For some, the long-term strategy may be to scale up and sell to Amazon or cash out in a stock offering. Others, according to Bell, seem driven to build their firms and remain in control. While Quidsi and Zappos agreed to be acquired by Amazon, he notes that Groupon rejected Google's $6 billion offer. "It comes down to the vision of the individual founder and the size of the opportunity to be had."

Dunn, the Bonobos founder, says service-driven, product-depth players like Zappos and Quidsi are good candidates for acquisition by larger firms. "We don't want to compete with them at their own game, which is low-cost product depth," Dunn notes. "In the long run, we would be more honored to be their competitor than their subsidiary. But we've got miles to go to even think about that possibility."

Not all entrepreneurs are drawn to the online world. Patrick FitzGerald is working on another diapers-oriented venture, Nanny Caddy. The company provides vending machines stocked with diapers and other baby items, such as pacifiers, wipes and sunscreen, to child-oriented locations, including zoos and museums. FitzGerald's prior start-up, RecycleBank, provides incentive programs for recycling. "You can't get more in the physical world than trash and vending machines," says FitzGerald. The key difference between managing an Internet and an offline company is the relationship with customers, which he argues is more powerful in a company that works primarily in the physical world. The downside is that a business like Nanny Caddy requires significant travel and hands-on involvement. "The logistical component of moving Nanny Caddy and delivering nationwide is not so easy. I have much bigger forearms than I did before."

FitzGerald stresses that first-hand feedback on the ground is critical. "You learn very quickly where the holes are," he says. "It's easy to sit in a room and pound away at spreadsheets, but sometimes you have to get your hands dirty."

Online or off, retailers will always face the pressure to generate profit, Wainwright notes. The online retail marketplace has a lot of "clutter," and e-commerce companies face challenges, she adds. "The economics are not that much different than bricks-and-mortar stores.   The margins are slightly better, but they are still thin margins."

Wharton's Hoch also adds a note of caution: "In the future, there is a lot of opportunity for start-ups in specialized e-tailing. But it's also easy not to succeed. The barriers to entry are low, and the barriers to exit are low. A lot of people are trying things. Sometimes you just can't tell exactly what is going to work."

*[A version of this article originally appeared at Knowledge@Wharton on May 11, 2011]