Lies, damned lies and fresh TLD statistics

Dig into statistics to understand how valuable they are.

Fresh top level domain name options have been out for less than a year, and everyone is clambering for data to make sense of the rollout.

That includes me. I look at registration data from a number of sources, weekly sales information, usage data…basically, anything I can get my hands on.

Yet it’s notable to understand that data can be misleading, especially if you don’t dig into the details. A prime example is looking at raw registration numbers as a sign of the popularity of a fresh TLD. These numbers are not always indicative of popularity since of freebies and registry registrations.

Here’s another data point that might not be all it’s cracked up to be: the number of registrations for any given TLD in the Alexa top million.

Even ahead of fresh TLDs, Alexa has been a notoriously poor ranking tool for websites. The fresh TLDs at the top of Alexa rankings are a case in point of how poor it is.

Did you know there are 43 “websites” in the top 25,000 on Alexa that are based on fresh TLDs? That’s a better ranking than any of the domain blogs, and these domains have been registered within the past year. In fact, some of these domains in the to 25,000 have been registered within the past month.

DomainPunch, which offers tools for tracking fresh TLDs in the top 25,000, decided to investigate how this is possible. Here’s what it discovered:

• 90% of these sites in the top 25,000 are either parked or redirected to an advertisement page or have illegal content.
• 26 of them (60%) are “.XYZ” sites. Out of these, 20 have the same IP address.
• All XYZ sites, except one, serve advertisements or are parked with no actual content.

Interestingly, many of the domains are registered through a registrar affiliated with the backend registry of the names.

I tested a number of the domains in the top 25,000 and most resolved to parked pages or were forwarded via zero click (counting one to a noted fishing site).

Call me crazy, but I’m pretty sure that ToesBait.xyz is not a more popular site than Reglike.

This is just another example of how you need to dig deeper into statistics to understand how valuable they are.