Will more contested TLDs get more registrations?

Probably, but quantifying it is difficult at this point.

I commonly hear that the best is yet to come in terms of new to level domain name launches. The most heavily contested applications are just getting resolved, and these represent the best top level domain names.

It makes sense to me. But will this reflect reality? Are the TLDs we’re seeing so far not representative of the results of future, heavily contested TLD launches?

I tried to dig into the data a bit this week, but I’ll admit it’s difficult to draw any conclusions. Among the factors that differ between each TLD that are difficult to control for:

1. Retail price
2. Number of premiums and their prices
3. Number of domains on name collision list
4. Registrar distribution
5. Date released
6. Marketing

The closest you can get to controlling for these is to compare Donuts’ domain names. You can at least eliminate #4 and #6, because most of Donuts’ domains are available at the same registrars and have been marketed the same way.

You still get into issues with retail price, premiums, collision lists, and dates.

The problem with dates is that most of the TLDs that were in contention sets are just now being released. It’s so soon after release that even an extra month on the market can make a difference in the registration base. A long time from now this difference will be marginal and you can run a better analysis.

I looked at Donuts domain names that ended sunrise in August through November, checked if they were contested, and then looked at the most recent zone file data from Domain Incite Pro (as of Thursday). This is what I found.

This list is ordered by when the domains were released. As you can see, more contested domains are coming online recently.

I’ve played around with the data and found that it’s really hard to draw any conclusions. The median number of registrations are higher for the contested domains despite most of them coming out recently. But we’re talking about small numbers here, and you still have so many uncontrolled variables. For example, .CreditCard is an expensive domain to register. Some of the domains clearly have broader appeal than others, too. To be sure, even the ones with 3-4 applicants aren’t exactly blowing doors.

Qualitatively, some of the contested ones look better than the uncontested ones. .Restaurant and .Church come to mind. Yet other ones, including the four-applicant .pizza, have fairly limited market size.

There’s another variable that gets introduced when there’s just one applicant for a domain: varying levels of quality that are difficult to quantify.

I think people are generally correct when they say the best is yet to come with new TLD quality. Yet I also question how much that will correlate with registration numbers. If you pay more for the rights to a TLD, you might hold back more premiums or charge higher prices.

Let’s check in on this in another year.