From: "James M. Bladel" <jbladel@xxxxxxxxxxx>
To: "PEDNR " <gnso-pednr-dt@xxxxxxxxx>
Date: Wed, 8 Dec 2010 08:13:32 -0500
Subject: [FWD: RE: [gnso-pednr-dt] Renewal Rates & Grace Periods]
Sorry, I had intended for this to go out to the group list.
Thanks--
J.
-------- Original Message --------
Subject: RE: [gnso-pednr-dt] Renewal Rates & Grace Periods
From: "James M. Bladel" <<mailto:jbladel@xxxxxxxxxxx>jbladel@xxxxxxxxxxx>
Date: Wed, December 08, 2010 6:11 am
To: "Alan Greenberg"
<<mailto:alan.greenberg@xxxxxxxxx>alan.greenberg@xxxxxxxxx>
Alan:
I would caution against drawing your first and last
conclusions. My data request focused on the time period -after-
expiry, so it is not clear what periods well before (e.g. months or
weeks) or well after (redemption) are not covered by this data. It
is very likely that large portions are not accounted for by this sample.
Similarly, I was not specific about normalizing renewal behavior for
names that "went dark" or were kept active (perhaps by other
products). It is possible that an associate service renewed even
earlier (e.g. the previous month(s)), or expired separately.
We are looking at a very narrow question here (use of renewal grace
periods), and trying to read more in to this data is
speculation The bigger picture is more complicated, involves
additional factors, and is probably one reason why registrars are
uncomfortable publicizing data "snapshots."
J.
-------- Original Message --------
Subject: Re: [gnso-pednr-dt] Renewal Rates & Grace Periods
From: Alan Greenberg
<<mailto:alan.greenberg@xxxxxxxxx>alan.greenberg@xxxxxxxxx>
Date: Tue, December 07, 2010 11:33 pm
To: "James M. Bladel" <<mailto:jbladel@xxxxxxxxxxx>jbladel@xxxxxxxxxxx>
James, I would like to make sure I understand these stats. To take
May as an example, the way I read this (rounding to nearest percent):
- 32% of expiring domains are renewed prior to or on the expiration
date. Therefore 68% of expiring names are NOT renewed prior to on
expiration. I find that disturbingly high.
- An additional 25% are renewed the next day. My recollection is that
it is GoDaddy practice to to redirect the web site at either day 3 or
5, so this is before it stops working as before expiration.
- only 4% are renewed just before and for several days after the
domain is redirected.
- The numbers add up to 63.5, so 36% of expiring domains were not renewed.
Am I reading this correctly?
At 08/12/2010 12:02 AM, James M. Bladel wrote:
>Team:
>
>Sorry I cannot join you tomorrow morning, but I have a previous
>commitment to the DNSSEC event.
>
>After some discussions this weekend, I spoke with our internal stats
>team and requested six months' data on renewal activity for
>gTLDs. Renewals were bucket-sorted by post-expiry time (<0 days, 1
>day, 2-10 days, 11-30 days, and 30+ days). The percentages are
listed below.
>
>
>(NOTE: I am not authorized to release raw numbers, but the average
>monthly renewals for this period was more than 2.5 million domain names.)
>
>
>
>
>This data indicates that a significant portion of names renew either
>before, upon, or within 1 day of expiration. Also noteworthy is the
>significant drop off in renewal activity after Day 2. Finally, it
>is clear that renewals are rare beyond the 10th day.
>
>In an earlier conversation, some of us (myself included) speculated
>that we might be "chasing the tail" of renewal activity for periods
>longer than 5 or 10 days. But even I was surprised at the
>substantial lack of renewal activity only a few days past expiration.
>
>Therefore, I submit to the team that the benefits of a longer grace
>period (>10 days) are negligible, and do not warrant the disruption
>to existing business practices and industry expectations.
>
>Thanks--
>
>J.