Applicants for the recently approved launch of new generic top-level domains (gTLDs) should seriously take into account niche markets instead of focusing on the number of registrations.

Drawing on failure and success drivers when it comes to Internet start-ups, it becomes clear that the failures of recently introduced generic top-level domains (gTLDs)—such as .biz and .info—are due to their focus on the number of registrations (i.e., scale) instead of focusing on niche markets. On the other hand, start-up successes have been consequences of prudent initial niche strategies.

I first outline two examples of start-up failures that focused on scale rather than niche markets. I then outline successes of two niche market strategies. The essay does not address gTLDs of brands, such as .canon.

One example of scale failures is FreeMarkets Inc., founded in 1995, which offered to save companies on their purchases through online auctions. It offered its clients a broader range of suppliers and provided information on the latest price reductions. The success of FreeMarkets led to a December 1999 initial public offering (IPO) leading to a market valuation of $8 billion at end of first day of trading.

Such a phenomenal success did not go unnoticed by FreeMarket’s major clients. General Motors, which accounted for about 17% of FreeMarket’s revenue, announced the formation of a consortium with rivals Ford and Daimler-Chrysler merely months after the IPO. The new entity, Covisint, would go beyond publishing price information and more into information-sharing among its members and suppliers. However, the cost of harmonizing the interfaces among all the manufacturers and hundreds of suppliers outweighed any benefits of economies of scale. FreeMarkets later acquired Covisint in 2003 and FreeMarket was sold in 2004 to Ariba Inc. for a fraction of its IPO price.

A second example of failure due to a scale strategy is Value America Inc., which tried to sell anything and everything online. It was founded in 1996 and funded by heavyweights such as FedEx founder Fred Smith and Microsoft cofounder Paul Allen. Its market value at the end of its first trading day was $2.4 billion. It filed for bankruptcy after only 16 months.

You might say that Amazon is an example of the success of a scale strategy. However, looking at Amazon now, their strategy was still that of a niche market as a “pure play” online bookstore. The company expanded through investments in strategic competencies and into adjacent markets. Nevertheless, brick-and-mortar bookstores that expanded into the online space have not fared well.

Zappos is another example of the success of a niche strategy. Founded in 1999, Zappos focused on shoes with an emphasis on attracting and retaining loyal customers. It was sold to Amazon in 2009 for $1.2 billion despite Amazon having its online shoe business. Zappos still dominates the online shoe market, which is estimated at a worth of over $4 billion.

In the domain name space, existing domain names such as .biz and .info have an unfocused message and are scale-based, because any business can use them without providing a clear message. Their strategy has been, “If we build them, they will come.” Nevertheless, you can compete with .com with a product differentiation strategy and a unique marketing strategy.

Niche strategies include focusing on location (such as .nyc for New York), community (such as .arab), niche industries (such as .cars). Product differentiation does not imply a niche strategy, but a viable niche gTLD message involves product differentiation. Applicants within a “similar” niche should compete on customer service, marketing, restricting openness, and bundled TLD services. Keep in mind that price competition can easily lead to destructive wars resulting in losses for all parties.

Although incumbent registries may have a competitive advantage in managing operating costs and in offering a wider scope of TLDs, they may not have marketing competencies for new niche markets. So, don’t let incumbents sway you away from new niche gTLDs when you have superior competencies.

Thus, applicants for the recently approved launch of new gTLDs should take niche markets very seriously. Niches include location, community, and industry signaling.