Summary

  • ICANN’s 7 October snapshot reports regional counts for both applications and applicants. North America leads both tables, but its share is 53.5% of applications and 41.2% of applicants; Asia/Australia/Pacific moves the other way, from 13.5% to 32.4%.
  • The figures describe filings and listed applicants. They do not establish regional representation, equal access, outreach reach, public support, or the outcome of later evaluation.

The denominator changes the map

The 2026 new-gTLD round now has a public geographic snapshot. ICANN’s 7 October announcement counts 1,615 applications across five regions. The companion Application Publication and Statistics site reports 481 applicants and breaks those applicants down by region. Read separately, each table is accurate. Read as though they counted the same thing, they tell a distorted story.

Region Applications (% of 1,615) Applicants (% of 481)
Africa 16 (1.0%) 7 (1.5%)
Asia/Australia/Pacific 218 (13.5%) 156 (32.4%)
Europe 506 (31.3%) 111 (23.1%)
Latin America/Caribbean 11 (0.7%) 9 (1.9%)
North America 864 (53.5%) 198 (41.2%)

The largest gap is not a contradiction in the data. It is a difference in units. North America accounts for more than half of applications, while a little over two-fifths of applicants are listed there. Asia/Australia/Pacific has a much larger share of applicants than applications. An applicant may file more than one application, so a filing count can rise faster than a count of participants.

That distinction matters to anyone describing the round as a global market. A headline built from 864 North American applications can suggest a stronger regional presence than the 198-applicant table does. Conversely, the 218 applications from Asia/Australia/Pacific do not capture the 156 applicants behind that count. Neither measure should replace the other: one describes volume; the other describes the number of applicants.

A count is not a constituency

The statistics still leave important questions unanswered. They do not show how many organizations in each region could have applied, how many learned about the round, how many began but stopped, or what costs shaped the decision. Nor do they identify the people who may eventually use a successful string. Applicants are not a sample of Internet users, and an applicant’s listed region does not authorize it to speak for everyone there.

This is a limit on interpretation, not evidence of misconduct or a failed program. The figures do not prove that applicants faced unequal barriers. They also cannot establish that the distribution is fair. That would require a defined objective and evidence matched to it, such as regional eligibility, outreach, costs, application progression and eventual outcomes. The current snapshot supplies none of those denominators.

The date matters as well. The announcement says its figures reflect primary-string application data as of 7 October. The APS site says it will not update again until String Confirmation Day. Applicants may select an eligible replacement string through 21 October; the final list and revised contention sets are scheduled for 17 November. Replacement records can appear separately on the application list while counting as one application in the statistics. The 1,615 figure is therefore a dated application count, not a final inventory of unique strings or an estimate of delegated registries.

The useful public record keeps both maps visible and labels each one. It should preserve the date and denominator when the next snapshot arrives, then show how applicants and applications move through replacement, evaluation, public input, objections and later contracting. A reader can then distinguish a larger filing footprint from a broader applicant base without treating either as a vote.

Sources: ICANN’s 7 October Reveal Day announcement; APS application statistics; APS update schedule; APS applications and replacement records; APS publication guidance; ICANN’s 29 September milestone announcement.