Why did ICANN commission this study?
As part of the Competition, Consumer Trust and Consumer Choice (CCT) review of the New gTLD Program, the ICANN Board adopted 66 recommended metrics from an Implementation Advisory Group made up of ICANN community members. A subset of three were identified from the 66 metrics as best being measured by an independent economic study. The ICANN Board adopted the recommendation to commission an economic study in March 2014 at the ICANN51 meeting in Singapore with the aim of establishing a baseline of competitive effects in the domain name marketplace. After a request for proposal process, ICANN signed a contract with Analysis Group to conduct the study.
The metrics included in the February 2015 Phase I study capture how competition has been affected in the domain name marketplace before and after the introduction of new generic top-level domains (gTLDs). The Phase II study, published in September 2016, allows for comparisons with the Phase I findings published in September 2015.
How was the sample of gTLDs constructed?
In Phase I, Analysis Group constructed a sample of 109 new gTLDs whose registration figures and pricing data would be included in the study; 14 legacy gTLDs which existed prior to the expansion; and 15 country code TLDs (ccTLDs) to ensure geographic diversity. However, the majority of ccTLDs did not provide wholesale data. Registration volumes for ccTLDs also were not available.
In Phase II, Analysis Group added 30 new gTLDs to the sample to account for registration activity in the year since the Phase I study was conducted. The Phase II sample included 139 new gTLDs and 14 legacy gTLDs based on registration reports for October 2015.
The TLDs were selected using a multi-step process:
- First, gTLDs with historically higher registration volumes (identified using Monthly Transaction Reports) and those available for purchase were included in the Phase One sample. This includes 14 legacy gTLDs which existed prior to the New gTLD Program: .COM, .ORG, .NET., .ASIA, .MOBI..BIZ, .INFO, .NAME, .PRO, .ASIA, .TRAVEL, .JOBS, .CAT, .TEL AND .XXX. Eight legacy gTLDs were excluded because they were reserved for specific entities with restrictive registration requirements: .GOV, .EDU, .MIL, .INT, .AERO, .COOP, .POST, and .MUSEUM.
- Analysis Group then included those new gTLDs with the highest registration numbers as of November 2014 and those with the highest registrations for the period September to November 2014 in the Phase I sample. The Phase II sample expanded the Phase I sample to include 30 new gTLDs that had the highest registration figures as of October 2015.
- Any new gTLDs similar in name and likely purpose were added to allow for comparisons between gTLDs likely targeting the same customer base. For example, if .WORK had been included, Analysis Group would have also included .CAREER, .JOBS, etc.
- The resulting new sample was reviewed to ensure it included “community” strings and geographic diversity. Within the sample, there are at least five Internationalized Domain Names and at least five community strings. In addition, three ccTLDs per each of ICANN’s five geographic regions were selected to ensure sufficient geographic diversity. The three selected ccTLDs reflected the highest number of domain name registrations, which were open for registration.
Which TLDs were included in the sample?
The exact TLDs included in the sample, with the exception of those that existed prior to the New gTLD Program, will not be disclosed to preserve the confidentiality of pricing information.
Can I see the data set?
No. In order to preserve confidentiality of registry and registrar pricing data, only Analysis Group has access to the complete data set. ICANN did not collect and does not have access to any of the confidential pricing data used in the study. All data presented in the report is aggregated and anonymized to the extent possible.
How was pricing information obtained for the gTLDs included in the study?
Analysis Group requested and directly collected sunrise and wholesale pricing data from the registry operators in both phases of the study. While some legacy TLD registries provided data, most data on historic legacy TLD wholesale prices was collected from official price change correspondence between operating registries and ICANN, available here. The majority of ccTLD registry operators did not provide wholesale data.
Retail pricing was more challenging to obtain in both phases of the study. Requests for current and historical data were sent to all registrars in the sample. Only six registrars, all from the Asia Pacific region, provided some form of historical data in Phase I. In Phase II, only five registrars, again, all from the Asia Pacific region, provided some historical data. These responsive registrars accounted for only 14 percent of registration volume of the new gTLDs being sampled in Phase I and did not provide any regional geographic variation. Consequently, Analysis Group collected posted retail prices from the sampled registrars’ websites in both Phase I and II. However, many registrars in the original sample (which was based on registration volumes of new gTLDs) did not offer publicly-available pricing information. As a result, retail price information was collected from 39 of the original 54 registrars in the sample for Phase I. In Phase II, if available, Analysis Group collected retail price data from Domain Name Prices for registrars and TLDs in the sample. When not available from Domain Name Prices, Analysis Group manually collected retail price data from the websites of registrars. As a result, retail price information from a total of 39 registrars is included in Phase II. Fourteen were available in the Domain Name Prices data and twenty-five were manually collected. Analysis Group also used registrars’ websites to collect data on add-on services like hosting, email, forward services and others.
In addition, Analysis Group received no data from secondary market institutions. Such data would have allowed for better investigation of how consumers value different domain names at legacy TLDs and new gTLDs.
How did Analysis Group obtain registration volumes?
Publicly-available transaction reports for each TLD, which provide information on historical registration volumes, were collected from ICANN’s website. These reports detail how many registrations each registrar was responsible for in each month. Historical registration volume data was unavailable for ccTLDs.