Home / Blogs

New TLD Launch: Lessons Learned

Tobias Sattler

In September 2017, I wrote an article [1] about the new domain extensions in German and got very good feedback and was asked to translate it into English in order to make it available for a broader audience. I wanted to comply with this request, but unfortunately, it took a while to revise and translate my article.

In June 2011, the Internet Corporation for Assigned Names and Numbers (ICANN) gave the starting signal for about 1,400 new top-level domains (TLD), such as .blog, .club, .cloud, .ltd, .nyc, .online and .shop, to make the existing namespace bigger and more diverse.

Far Too Much New GTLDs at Once

After a lengthy preparation phase at ICANN, more than 1,200 new domain extensions have been introduced step by step since the beginning of 2014. Until the end of 2018, we will probably see another 100 to 200 TLDs starting.

With more than 23 million registered domains [2], across all new domain extensions, and the calls for a new round to accommodate more top-level domains, it is worth looking back to review the introductions and take a closer look on what we can learn from it.

Before the launch of the new TLDs, there were already 248 country-specific and 20 generic domain extensions. As a result of the introductions, the namespace grew from 268 to over 1,500 extensions within two years — truly an explosion. While it may be argued that some new endings are meaningless or nonsense, it has been rather the short period of two years that has led to the new domains having a harder time than many industry insiders expected or had wanted. Also, due to the sheer amount of new TLDs, it has been and still is, a tough job for the domain industry's players to draw attention to the new endings without overstraining their own customers and the public.

High Complexity on the Introduction of New gTLDs

Numerous new domain name registries with few experience, new business ideas, and new industry guidelines were making it even harder to promote the acceptance and distribution of the new domain extensions. Choosing some very restrictive conditions and one-sided contracts, some new registries already blew their chances even before the launch of their new TLD.

If we take a closer look at the start of a domain extension, then ICANN requires a so-called "Sunrise period" of at least 30 days during which, trademark owners can pre-register domains that correspond to their registered trademarks to protect them from IP-right infringements. For this purpose, the Trademark Clearinghouse (abridged TMCH) of ICANN was originated. This added a further complexity and was exploited by word and figurative marks, considering that there were also registered trademarks for Berlin, pizza, cloud, storage, etc. ICANN will have to take a closer look at future rounds in order not to lose sight of the true purpose of the TMCH.

After the Sunrise period usually follows the Landrush phase, which was mostly 30 days long in previous TLD launches. However, it was hardly used. Many registries took advantage of the freedoms granted to them and introduced a new preferred registration, the Early Access Phase, at much higher costs.

Depending on the domain extension, this period lasted 5 to 20 days, and thereby one-time costs incurred, which were lowered following the Dutch auction model for the respective domains per day until general availability has started.

Compared to previous TLD introductions, now more registries utilized the option of premium domain names. These were, as in previous introductions, distributed by the domain registries in part directly in cooperation with an auction provider and, on the other hand, also directly via domain registrars and sold individual domain names at a higher price if the domain had a perceived higher value to the owner, public, etc.

This sometimes lead to different processing and pricing models for the TLDs, which had to be explained to potential domain buyers. In addition, a combination of premium domains within the Sunrise, Landrush and/or Early Access Phase was possible, which further increased the complexity and the communication effort.

In my opinion, in an ideal case to get a top-level domain up and running without overexerting potential domain buyers and domain registrars is a 30-day Sunrise, an Early Access Phase of 5 to 7 days, a manageable number (less than 5,000) of Premium Domains sorted into a few price categories (less than 6) and selling through brokerage for very pricey ones.

Technical and Operational Challenges Due to Non-Scalability

Technically speaking, the Extensible Provisioning Protocol [3] (abbreviated EPP) is available for registering domains, which, as the name implies, is extensible. Due to the different approaches of the domain registries, the lack of best practice and the sometimes lack of clear and binding guidelines on the part of ICANN, this led to a variety of implementations that the domain registrars had to implement and, in conjunction with the sometimes restrictive conditions, also had to further develop their own business processes. Gradually, an EPP extension for premium names [4] seems to have crystallised as a standard, although many different versions of it are in use.

Domain registries manage one or more domain extensions, either themselves or in collaboration with a backend provider, sometimes different registries even use the service of the same backend provider. Due to contractual regulations, lack of trust and few experience with administration and operation of a TLD, this led to separate systems per TLD, despite the same backend provider. For registrars, this created an unnecessary extra effort, which could have been avoided with sufficient coordination, planning and foresight from the registries.

Almost every TLD often has more or less maintenance work during the year. During this time, use of EPP may still be possible, but the administration may be limited. The time windows in which such work takes place are all too often communicated too short in advance, which can lead to overlaps with registrar marketing campaigns or the registrar's internal maintenance efforts.

A machine-readable announcement of maintenance announced at least 30 days in advance, would greatly facilitate the lives of everyone involved. Regardless, it would, of course, be better if it did not come to such maintenance necessities in the first place.

Moreover, the management of premium domains and reserved names, depending on the respective registry, is a complicated endeavor. Unfortunately, up-to-date lists are even today not available for registrars aiming to sell the premium domains for the registries, which means that during a maintenance work, a domain may appear as taken, even though it would be available because the status or price could not be clearly determined. Leading to a lower customer take rate, which no-one can be satisfied with.

By now the ICANN registry and registrar stakeholder groups are working together in a new formed TechOps sub-group to address technical and operational needs and challenges. The future will tell if they will be able to fix these issues, but I am confident that bilaterally, many of those challenges can be solved.

Marketing and Awareness – the Assumptions Were Wrong

With most of the new domain extensions launched, one could assume that the work is done, but the domain business is also characterized by constant change and customisation. Contractual changes of the registries or the sale of a TLD to a new operator and the concomitant changes are always expected to bring new challenges.

Relatively high ICANN and consulting costs for the acquisition of a TLD, possible legal disputes with other competitors for a domain extension as well as pressure from investors, meant that many registries soon had to reach break-even. The results are high domain prices and many premium domains.

The success of a TLD, however, is determined by general market acceptance and effective end customer marketing and sales initiatives.

Many registries, especially the younger ones, relied heavily on the fact that the domains would sell themselves by the hype of the new extensions. However, the hype, if you want to speak of such, was given only within the industry. Beyond that, many people are unaware of what a domain is, let alone that there are over 1,500 extensions today.

For registries, domain registrars are the only major channel of distribution, and they assumed that they would accept the new endings without any ifs and buts and also take over marketing. However, due to the numerous launches of new TLDs, registrars were more or less able and reluctant to choose for the first time if they want to carry one or the other domain extension, as it were either not relevant or too expensive for their markets, too restrictive or contractual terms are too one-sided.

Some registries came up with the idea to distribute domains in large quantities for free without any restrictions. As good as this idea may be, all the examples I know of did not work out because giving away domains for free will always attract the wrong people and this will ultimately harm the TLD. Therefore, it is better to have some restrictions, such as having to attach a service to this free domain.

Meanwhile, ICANN is constantly confronted with the demands that they should finance marketing and awareness campaigns from the money they have earned by introducing the new domain extensions. On closer inspection, this seems a difficult and in my view almost impossible project. Due to the principle of equality, ICANN would have to make sure that nobody feels disadvantaged within the community. This would very likely lead to long and numerous discussions in order to ensure a fair distribution of the money. Due to the different mixed situations, this would probably take several years  — with very small chances of success.

As if the topics of marketing and sales were not difficult enough, there were and still are some technical difficulties, such as internationalization and third-party adoption, with one or the other domain extensions, which also affects the market acceptance.

Unpleasant Pitfalls

The term Universal Acceptance (UA) is a concept for the equal treatment of all domains, with the aim of enabling Internet users to use the Internet in their own language. The use of internationalized domains, e.g., müller.de, nic.コ ム or новости.рф still cause problems for many websites, browsers and email programs. The assumption that domain names are written in ASCII [5] format has been outdated for several years, but there are still many applications that are unable to accept, validate, process, store, or maintain internationalised domains display. The Universal Acceptance Steering Group [6] was established to raise awareness on this issue and overcome the hurdles of using new domain extensions and internationalized domains. However, there is still a lot of work to be done two years after the foundation of this group until finally, all new gTLDs in local character sets work across the platforms.

Another example is the Public Suffix List [7], a project of the Mozilla Foundation. This is to make it easier for Internet browsers (such as Firefox, Chrome, Safari and Opera) to immediately recognize a TLD to make a distinction if an entered text in the address bar is an Internet address or a search term. In particular, at the start of the new TLDs, these were not part of this list or were added with a great deal of delay. As a consequence, new domains, e.g., nic.guru or nic.today, were not immediately recognized and therefore interpreted as a search term, which led to a search at Google or Bing instead of displaying the correct website. By now, the process has gotten better through closer collaboration between ICANN, registries, and volunteers from Mozilla, which is why these cases rarely or no longer occur.

Future Prospects

Despite all the difficulties surrounding the launches of the new top-level domains, they will find their way into everyday life and expand step by step. Even .com took some time to skyrocket. However, the consolidation of the domain market will continue and some top-level domains may disappear again.

Nevertheless, a lot looks like there will be more new TLDs from 2020 onwards. However, this time around, it will most likely be mainly trademark owners who will apply for their ending. However that may be, it is important to continue to work on the success of the new extensions, true to the motto "Keep it simple, stupid" (KISS), all parties should not lose sight of the big picture.

Feel free to reach out anytime. Twitter: tobiassattler / Facebook: sattler

References:

[1] https://medium.com/@tobiassattler/neue-domain-endungen-was-wir-als-internet-community-daraus-lernen-k%C3%B6nnen-71595444077e Retrieved 2 January 2018
[2] https://ntldstats.com Retrieved 2 January 2018
[3] https://en.wikipedia.org/wiki/Extensible_Provisioning_Protocol Retrieved 2 January 2018
[4] https://tools.ietf.org/html/draft-ietf-regext-epp-fees-08 Retrieved 2 January 2018
[5] https://en.wikipedia.org/wiki/ASCII Retrieved 2 January 2018
[6] https://uasg.tech Retrieved 2 January 2018
[7] https://publicsuffix.org Retrieved 2 January 2018

By Tobias Sattler, CIO united-domains
SHARE THIS POST

If you are pressed for time ...

... this is for you. 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.

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

Share your comments

To post comments, please login or create an account.

Related

Topics

DNS Security

Sponsored byAfilias

Cybersecurity

Sponsored byVerisign

New TLDs

Sponsored byAfilias

IP Addressing

Sponsored byAvenue4 LLC

Domain Names

Sponsored byVerisign