The new gTLDs program can’t succeed unless two things happen. The approved registries must do good, and ICANN must weed out applicants who are in love. This is to say that registries should put users’ good first, and applicants shouldn’t get the nod unless their motive is economic and/or social viability.

A recent study reveals that leading companies have enjoyed healthy profits because they made doing good their strategic foundation. The authors analyzed the core values of 34 companies from Fortune’s list of the 50 most admired companies. Each of the 34 was on the list for the past five years running. Each company’s values, as stated on its website, drives the company’s’ mission, strategy, and vision. Stakeholders know when a company deviates from its values, the result is downward pressure for stock price and management.

By discourse and action, new gTLDs and the registries have failed to put customers first. They have ignored the importance of doing good.

When it comes to discourse, we’ve heard about product and pricing-mechanism differentiation, and about the number of registrations (with a few strategies that included free giveaways). We have not heard about an integrated approach and doing good. And too often the registries engage in dubious rhetoric about their gTLDs while having nothing to say about the real risks and costs of rebranding.

As for action, the registries have made no attempt to limit phishing, among other problems. Nor have they been proactive about designing a mechanism for trademark protection. Brand protection should have been framed more broadly as doing good or putting customers first, rather than passively supporting limited listings in the Trademark Clearinghouse (TMCH).

For the next round of new gTLDs, registries should limit their reliance on first-round success factors because:

1. There is no agreed-on metric for quantifying success.
2. Profitability does not necessarily imply an optimal or a satisfactory financial performance.
3. When new gTLDs make a profit, it is not clear what the driving forces were.
4. Past success doesn’t predict future improvement. People who see or experience a winning streak often assume that the performance will keep getting better, and they make decisions accordingly. A recent study finds evidence that these people are making a mistake. The higher a participant’s initial score, the less likely their score was to improve later on. Success may be a negative predictor of future performance improvements, the study indicates.
5. The investment returns will be greatest from picking the right emerging product gTLDs. Of course, such investments come at higher risk. The selection process should be based on analyzing emerging industry labels. It shouldn’t become a guessing game as to what would be a successful gTLD.
6. Selection flops from the first round do not necessarily disprove the possible benefits of the new gTLDs program. Recent studies and company successes reveal the disadvantages facing first movers (in this case, first-round selections), as pioneers can make costly mistakes.

ICANN should try to limit new gTLD applications whose social cost is greater than their benefits. Given that the market (and not beauty contests) should arbitrate what is valuable, there are still mechanisms that can make the market more efficient. For one thing, ICANN can approve only those gTLDs that receive multiple competing bids (a sign that more than one entity believes a given gTLD will create value). This isn’t a perfect solution, but it improves the chances of ending up with new gTLDs that bring net social gains.