I’ve been asked several times about Paul Stahura’s blog post in CircleID. Long story short, Paul Stahura of Demand Media claims his analysis proves that “there is no economic incentive for an applicant to obtain a TLD for the sole purpose of making money from defensive trademark registrations.” This reads as a rebuttal to concerns expressed by the intellectual property community regarding the potential launch of unlimited new TLDs.
 
I’ve taken a look at the post and the analysis appears to be based on some deeply flawed assumptions. The research does not reflect the reality of registration costs or domain name strategy, and as a result cannot be used to project what will happen in a newly expanded TLD space.

 
First of all, the “costs to brand owners” for each domain is calculated with the $8 registration fee that most wholesale-market registrants are charged. The fact is that brand owners are charged substantially higher prices for domain names during TLD sunrise periods. Some sunrise-period schemes in recent years have called for fees as high as $10,000 per domain. Registries know that trademark owners will have to choose between either paying up front for domains containing their mark or risk having a third party pay just $8 to purchase it later; they also know that faced with these two choices, brands are likely to pay the high prices to ensure the domains that are most important to their companies are secured in advance.
 
It seems that whenever the domain industry needs a revenue boost, domain industry insiders use the ICANN policy development process to spawn new TLDs and prey on trademark owners’ fear of infringement by giving them the option of pre-registering their brands at inflated prices. Brand owners feel blackmailed, and even when they do register hundreds and thousands of domains, there is no way they can secure every possible combination of their brand in the new domains.
 
Second, no one has said that brands are going to go out and register in all the new TLDs that are going to be launched, so checking to see how many of the same domains are registered across dot-COM, dot-NET, dot-ORG, dot-INFO, dot-BIZ, dot-US and dot-MOBI (which, quite frankly are not the most popular TLDs and almost completely leave out frequently used geographic TLDs) is not a test that will help predict behavior in a new TLD space. If brands register defensively, they will do so in the TLDs that make sense for their business model and will be forced to pay dearly for it.
 
Third, one should not only measure how many trademark owners are using the sunrise and how many of the same domain names are found across multiple TLDs, but what sort of content is found on these domains. For example, if the content is the same across all TLDs, or if domains don’t resolve, then it is a sign that the registration is part of the brand’s defensive strategy. If the content is different, then it is somehow part of the brand owner’s promotional strategy. Only by understanding these differences can one draw conclusions and make predictions about a brand’s current and future strategy. Paul’s research fails to take these differences into account.
 
Fourth, neither squatters nor brand owners would register hundreds of misspellings in extensions other than dot-COM. Misspellings are about user experience, and since users are programmed to type in dot-COM (for the most part), registering many typos in other extensions would be useless. Paul’s decision to analyze 263 domains containing the misspelling of “Verizon” across 7 TLDs really cannot produce relevant conclusions.
 
All of these faulty assumptions and faulty research lead to faulty conclusions. From his research, Paul deduces that we can expect about 7,000 sunrise registrations in TLDs. However, we know that 15,334 applications for domains names were received by the time the dot-ASIA sunrise concluded, and 245,000 domains were applied for in the dot-EU sunrise. Speaking on behalf of the brand community, while Paul’s time and interest is appreciated, the four major observations above disprove his argument that brand owners are not actively protecting their brands in the various contemporary TLDs and that an onslaught of new TLDs will not cost the global brand community a massive amount of unrecoverable resources.
 
To get a better idea of what can actually happen in the new TLD space, check out the CADNA analysis of likely new TLD scenarios. This report places the estimated “haul” for registrars and registries from the brand community at $1.6 billion.