Home / Blogs

Metrics for New gTLDs Performance

The essay describes six groups of gTLD performance indicators: (1) the presence of a secondary market; (2) uses of second-level domain names; (3) the identity of the owners of parked and not-in-use domain names; (4) domain name renewals; (5) the market size of the supporting complementary assets; and (6) the character of the person in charge of a given registry. A time-trend of these indicators would convey any price inefficiencies and bubbles, and the health of the industry. However, there is no one metric or an easy combination to measure the indicators of performance of the new gTLDs.

One indicator of success is based on information derived from a secondary market for second-level domains: prices and transactions transparency. However, such a market can be a double-edged sword. It provides prices that reflect transaction volumes, who is buying and selling, and the average expectation of market participants as to what various domain names are worth. Unfortunately, the other edge of the blade may be a thinly trading market that reflects inefficient prices that can be manipulated by investors and can also lead to speculative bubbles. A good example of the association of fads and markets is described in a recent book by Zac Bissonnette, The Great Beanie Baby Bubble. The Beanie Babies mania lasted from 1996 to 1999. Bissonnette describes how eBay was instrumental in creating the market for Beanie Babies. EBay gave the market just enough transparency for potential buyers to watch as a Royal Blue Elephant, for example, zoomed to $5,000 at the craze’s height. Seeing prices jump like that contributed to the bubble, one where a line of 1996 election-themed Beanies, retired at the beginning of 1997, jumped in price to US$50 in two months and $625 a year later.

A second indicator is what the domain names are being used for. Yes, gTLDs are intended to be used by businesses for branding and labeling. Businesses can also use them for forwarding or as doorways, and investors can use them for parking; the names can also lie idle. The type of use is important; for example, the ratio of total business to total registrations can indicate whether investors find that the domains actually help operations or whether businesses hope to make money off the domains by reselling them. A high concentration of investor ownership can suggest speculation. Low business use is another sign, a disturbing one; we learned as much in 2008, when trafficking in complex derivatives of dubious economic benefits brought on a crisis. The Beanies taught the same lesson. The people driving up the Beanies’ price never thought that the things were toys and that children lose interest in toys. Speculators chased a killing while the dolls piled up in retail stores. The more an instrument is used as intended, the more it will be on solid ground.

The fourth indicator focuses on domain name renewals. There are a number of variables that should be looked at, individually and in aggregate. These include the ratio of renewal price to total number of registrations, the number of years of the original registration period, renewal prices compared to original prices, and whether the name was registered during sunrise or open registration periods. These indicators provide insight into the acceptance of the instruments for businesses and by investors. Registries need to model renewals, but it is not a trivial endeavor due to nonsynchronicity of registration and renewal dates.

A fifth indicator is the market size of the supporting complementary assets that gTLDs create—resellers, consultants, parking companies, hosting companies for the gTLDs and domain names, and so on. The composition of the services can be an indicator of the primary use of the gTLDs. With manias, it’s not just the direct participants (market facilitators, buyers, sellers, and registries) that benefit. According to some accounts of the California Gold Rush, “As the boom continued, more and more men got out of the gold-hunting business and began to open businesses catering to newly arrived prospectors. In fact, some of America’s greatest industrialists got their start in the Gold Rush.” As for investors, history tells us that most likely they won’t thrive. According to Dr. Gayle Olson-Raymer, a Humboldt University historian, “Very few individuals really prospered from placer mining. Most of the large profits fell into the hands of corporations who could afford hydraulic and quartz mining.” But owners of complementary assets may profit tremendously, at least judging by what happened with Beanie Babies and the Gold Rush. In the Gold Rush, they did better than the prospectors.

A sixth indicator is the character of a registry’s CEO. Researcher Fred Kiel has quantified this factor in his book Return on Character: The Real Reason Leaders and Their Companies Win. He classified as virtuosos those CEOs who achieved a 9.35 percent return on assets (ROA)—defined as net operating income as a proportion of total assets—while those tagged as self-focused scored just a 1.93 percent ROA. This was over a two-year period, i.e., more than five times. Unlike ROA, of course, character is hard to measure. Kiel and his research team decided that the word stood for integrity, responsibility, forgiveness, and compassion. Those four “universal principles” were then refined into key behaviors (such as “telling the truth” for integrity). And it was those behaviors that the researchers sought to measure. In a related study, Stanford University’s Charles A. O’Reilly III and his colleagues found that an organization’s culture tends to reflect the personality of its CEO and has a significant impact on firmwide performance. The authors find that certain personality types are associated with a wide variety of positive outcomes that enhance a firm’s status and prosperity.

By Alex Tajirian, CEO at DomainMart

Filed Under


Great article. I really think you are Colin Campbell  –  Apr 21, 2015 9:47 PM

Great article. I really think you are on to something here. It would be valuable if someone did a “MorningStar” style rating based on your criteria. Maybe NameStat.org or NTLDstats.com will take notice and develop such a list.

Tools not interpretations Trey Harvin  –  Apr 22, 2015 3:05 PM

Metrics are necessary tools for interpretation, but aren’t interpretations on their own.  You outline several valid ways to compare old and new TLDs.  But comparing and judging are not the same thing, particularly given the inherent differences in TLDs, old and new.  More subjectivity would be needed to judge success or failure. Here are a few examples:

* Compare these metrics along the life cycle to show velocity at sunrise, launch, pre-renewal, renewal, etc. Showing sales activity relative to movement in time would better highlight different business models and identify similar or dissimilar patterns to forecast future behavior.

* Group metrics for subjectively similar TLDs using what is known about their business models. Some will be concerned with large corporates, others with personal use, others with geographic concentrations, etc.

* Again based on business models, compare metrics with addressable market penetration for target buyers, weeding out parked domains and showing in-market velocity growing as marketing plans run their course (very valuable for domainers as well).

* Via some websurfing, compare penetration for each into distribution channels (using readily available registrar metrics) to show the % of presence in the target market, overall and by geography when rollouts are market by market in the early stages.

As for the 6th point, that evaluation would be very subjective depending on the evaluator’s point of view and is not realistically quantifiable despite its acknowledged importance. Kiel’s evaluations in his book are based on many interviews from a material sample size of often insiders, and current TLD evaluators will not have these insider insights into character. Outside parties bring their own filters, and truthfulness is not often what one side or another wants to hear. In unpleasant situations, stakeholders are often at odds, can be highly charged, and may rather see a single outcome than the truth. CEO politicians can make everyone believe erroneously that their outcome is being supported, while upfront CEOs take near-term heat for frank observations and decisions. The politician CEO would have the highest marks in a real-time evaluation, while it takes longer observations or post-mortem evaluations like Kiel’s to reveal honesty or lack thereof. Thus, this 6th point is aspirational, but not realistic to quantify without insider insight providing rationale and facts.

Colin, thanks! I am more than happy Alex Tajirian  –  Apr 23, 2015 4:57 PM

Colin, thanks! I am more than happy to collaborate with anyone interested.

Hello Trey,Thanks for your thoughtful and specific Alex Tajirian  –  Apr 23, 2015 4:59 PM

Hello Trey,

Thanks for your thoughtful and specific comments.

You are correct that “[m]etrics are necessary tools for interpretation, but aren’t interpretations on their own.” However, you did not suggest any interpretation techniques. The most reliable interpretations can be derived based on statistical models. (I believe that’s Colin’s point in the above comment.) I have used such models to examine the factors/metrics driving the market prices of domain names. Yet, before running such models, the researcher needs to visually examine various plots of the data.

Responding to your comments:
1.  Showing the “velocity” of sales at various points in a gTLDs’ life cycle can be easily handled by using a dummy variable reflecting its age. Based on the post’s title, I was not trying to compare old and new gTLDs, however, possible differences would be captured by the inclusion of such a dummy variable. You can also include other variables to distinguish personal and geographical gTLDs.

2.  It is not straight forward to measure “market penetration,” as it is not easy to determine the true boundaries of a market at any given time. Nevertheless, the implied gTLD signal can change over time, as we have seen, for example, with .net and .org.

3.  You are correct in noting that quantifying character is not trivial. However, you can use proxies for registries’ character, such as their concern with typo- and cybersquatting, as possible measures by their proportion to total registrations.
Nevertheless, my essay does not necessarily provide an exhaustive list of viable drivers of gTLDs’ success.

Comment Title:

  Notify me of follow-up comments

We encourage you to post comments and engage in discussions that advance this post through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can report it using the link at the end of each comment. Views expressed in the comments do not represent those of CircleID. For more information on our comment policy, see Codes of Conduct.

CircleID Newsletter The Weekly Wrap

More and more professionals are choosing to publish critical posts on CircleID from all corners of the Internet industry. If you find it hard to keep up daily, consider subscribing to our weekly digest. We will provide you a convenient summary report once a week sent directly to your inbox. It's a quick and easy read.

I make a point of reading CircleID. There is no getting around the utility of knowing what thoughtful people are thinking and saying about our industry.

Co-designer of the TCP/IP Protocols & the Architecture of the Internet




Sponsored byVerisign

Domain Names

Sponsored byVerisign

Threat Intelligence

Sponsored byWhoisXML API

Brand Protection

Sponsored byCSC


Sponsored byDNIB.com

New TLDs

Sponsored byRadix

IPv4 Markets

Sponsored byIPv4.Global