[gnso-rpm-wg] 99%+ reduction in sunrise utilization rate per TLD supports EFF call for elimination of sunrise

George Kirikos icann at leap.com
Mon Aug 14 12:26:29 UTC 2017

Hi Volker,

On Mon, Aug 14, 2017 at 8:02 AM, Volker Greimann
<vgreimann at key-systems.net> wrote:
> Thanks George. Having the list public serves many benefits, least of all to
> allow registrars that could not offer new TLDs until the claims phases
> closed to see these domains from day one simply because their systems di not
> allow for processing of these notices.

I agree with you 100%, that there are many benefits that should far
exceed the costs. It should have been public from day 1, and this PDP
has the opportunity to correct past mistakes, based on the
experiences/data/evidence that didn't exist before the policy was

One of the things I wanted to ask you (and other registrars and/or
ISPs can weigh in too), is whether you have any data on ARPU (average
revenue per user) for sunrise registrations, compared with 'normal'
registrations (e.g. non-defensive registrations, registered in
landrush or GA).

The reason I ask this is that it's an important consideration when
looking at the overall costs/benefits of the sunrise period.

My hypothesis is that a TM holder buying a domain for defensive
purposes *isn't* going to be buying things like webhosting, SSL, web
design services, email hosting and all the other kinds of offerings
from registrars and resellers (especially the consumer-oriented
registrars like GoDaddy, Tucows, etc., as opposed to the 'brand
protection' registrars like Markmonitor, CSC, etc.). In essense, that
sunrise-registered domain name is blocking someone else who *would* be
buying those high margin value-added services from registrars and
their resellers/partners. If my hypothesis is correct, that difference
should be reflected in the ARPU of registrars and their
resellers/service providers. It should be measurable.

We know that names 'captured' in sunrise have some  obvious negative
effects on registry operators. For exampled the delayed starts,
compliance costs,  being forced to spend on monopolistic providers
(TMCH etc.) and regulatory burden. There's the "deprivation of oxygen"
as defensive sunrise-registered names aren't 'used', and thus not
promoted by the end-user which would promote the entire extension
(e.g. just like Verisign doesn't have to spend much promoting dot-com,
because there's are many end-users already promoting active websites
ending in dot-com).

But, does that also hold true for the registrars, via experience/data?
I'm sure the registrars would guard the precise *magnitude* of that
data (except for public companies like GoDaddy which publish their
total ARPU). But perhaps some registrars might share the *relative*
ARPUs for the various subsets of registered names.

e.g. if a sunrise-registered domain name has a ARPU indexed to be 1.00
(or $X), is the ARPU for a landrush or GA domain name higher, or
lower, and by how much? (e.g. if it's 2.00 or $2X, then that's a real
cost of sunrises for registrars/resellers/partners).

This is perhaps data that should be broadly solicited from
registrars/hosting providers/web designers, etc., and added by ICANN
staff to the sunrise data collection document.


George Kirikos

More information about the gnso-rpm-wg mailing list