Featured Domains


New Top Level Domain registration numbers are increasingly misleading

Reporting aggregate numbers for new TLD registrations is misleading.

Some day this month one or more companies will proclaim that new top level domain name registrations have topped one million domain names.

It will be very misleading when they announce this.

If you look at the top 10 domains by zone files, at least three of them include a lot of free or registry-owned domain names.

The biggest example is .xyz. Network Solutions gave away tens of thousands of domains to customers at no charge. The customers didn’t even have to accept the domain; it was registered on their behalf.

In the case of .berlin people had to at least request the domain. .Link is also in the top ten, but I believe about 20,000 domains are registry owned and activated.

I’m not blaming all of these registries for being misleading. Frank Schilling never asked for people to report the top line from the zone on .link. He didn’t activate the domains to make it look like .link had a lot of registrations; he did it to create billboards on each one.

Smaller registries, including .buzz, also have premium domains activated in the zone.

Just how many domains in the zone files are paid and unpaid is hard to calculate.

On the flip side of this equation, companies that paid Donuts and Rightside to block their domains from registration aren’t counted, while you could argue they should be. Then you have the issue of the zone files undercounting the total number of registrations.

This makes it hard to really know when new TLDS collectively pass one million registrations. But I do know this: when that first company proclaims new TLDs have hit the million domain registration milestone, they’ll be wrong misleading.

DomainAgents. What should you sell your domain for? Read our Domain Market Report Now. Sponsored.

Get Our Newsletter

Stay up-to-date with the latest analysis and news about the domain name industry by joining our mailing list.

No spam, unsubscribe anytime.

Reader Interactions


    Leave a Comment

  1. Acro

    I say we count “intent to register” as well. As in, “hey Andrew, do you intend to register DNW.xyz?” – Yes. OK, that’s +1 for .XYZ. This will be called the “Domain Minority Report.”

  2. Philip

    I very much doubt the free or registry owned domains will contribute a significant amount to the overall 1 Million figure when announced.
    Sure, you’ll no doubt have a few tens of thousand component of that but that pales in comparison to the actual Million figure.
    Not something I think is a major issue as the overall figures are increasing all the time as more domains enter root.
    Will we be having this discussion when the figure hits 2M, then 10M or 50M. Only a matter of time for those milestones to be achieved I believe.

    • Acro

      The cumulative numbers are irrelevant. The introduction of gTLDs was meant to stimulate use, due to the (alleged) lack of .com availability. These little tricks and games in the initial launch are akin to pumping stock during an IPO. Registries that want customers and end-users should spend money reaching out to them: social media, direct sales etc. Otherwise it will take until 2050 to make a dent in .com etc.

      • Philip

        ‘Little’ is the operative word as a component of the cumulative number, hence hardly ‘irrelevant’. The actual, absolute number may not have a scientific degree of accuracy but is indicative at least of an overall demand for relevant domains.

        Let’s not miss the Wood for the Trees.

        • Acro

          What matters isn’t how many thousands or millions ALL gTLDs amount to. Each one is a separate product. Unless ICANN ran the gTLD ‘supermarket’ selling ALL gTLDs under one roof, the sales of bologna don’t mix those of milk and toilet paper. I hope that’s a good enough analogy.

      • John McCormac

        Good points, Acro. The problem was that there was never a “shortage” of .com domains. The domain names were being hoovered up on the drop by operators who would then taste them for PPC. Some domains would be kited from iffy registrar to iffy registrar so that they never really dropped and ICANN never got paid. When ICANN imposed that “restocking” fee, the bottom fell out of the business for many of these Domain Tasters. And ICANN had to be shamed into taking that action.

        The solution (for 2008) – more gTLDs. That’s the logic behind some of the new gTLDs – answering a question from 2008. Many of the new gTLD strings share characteristics with the keywords that were considered important and valuable during the Domain Tasting period. The problem for many of the new gTLDs is in their finite appeal. They are not generic as .com is but rather topic specific or niche. For ICANN, the market was shifting because the artificial shortage in dropping .com domains was fuelling the growth of ccTLDs. In countries where there is a strong ccTLD, the .com TLD has moved to being almost a legacy TLD and the highest growth and development is in the local ccTLD. This trend is beginning to impact .com now and growth is slowing down. While .com is still the top global TLD, it is losing ground at a local/country level. And these new gTLDs, like .xyz, may be awaiting drop day with some anxiety. But since one new gTLD registry has made such a deal, when the registration volume slows down, more will probably consider such a promotion. And it could spawn a bunch of mini .info gTLDs – bubble TLDs where growth is cyclically linked to promotions rather than interest.

  3. James

    Im interested in learning more about what happened with XYZ. Specifically, how “free trial” domains satisfy the RAA’s requirement for registrants to accept a Registration Agreement.

    • John Berryhill

      It doesn’t seem to comport with the requirement to obtain agreement to the terms of registration to have a “negative option” based on inaction.

  4. DonnyM

    Imagine a new company trying to compete against the iphone. That is what it is like for these guys.

    When google and amazon give away the domain names with the cloud and everything else does anyone really think down the line people will pay for a second tier domain name. These guys are the only ones that have money to promote domain names.

  5. Brad Mugford

    Despite radio silence from Mr. Negari, what has happened is fairly obvious at this point – these domains were assigned to accounts, free of charge, simply because they did not opt-out. That is clearly not providing explicit consent.

    I am not sure what kind of deal .XYZ and Netsol had, but the end result is worthless registrations than make up the vast majority of all .XYZ registrations.

    This looks like a blatant attempt to cook the books, and show higher demand than there actually is. XYZ registry got caught and now these sleazy tactics have been exposed. The domain world is pretty small, once your credibility is gone it is gone.

    ICANN cannot allow registries and registrars to engage in these type of tactics. There is no reason for a registrant to own a domain that they have no interest or intent in registering. This can lead to higher renewal costs and potential legal issues for the registrant in the future.


  6. Domenclature.com


    I am happy that bloggers such as yourself are scrutinizing these activities. If you recall, a few months ago, I asked you to explain the Zone file for your readers, which you did. The data collected from the Zone files should be analyzed Quantitatively, and Qualitatively.

    As the Registries focus on the Numerical (quantitative) data, looking at the mean, frequency distribution, standard deviations from the mean or confidence intervals, percentages and rankings, and higher level of statistical analysis, we will focus on the Narrative (or Qualitative) data. Since this type lacks a built in structure, it is going to be much harder to manipulate. Things such as Relevance, Effectiveness, Efficiency, Results/Impact, and Sustainability.

    We will be looking at the distinction between association and causation and the role that can be played by Confounding factors in skewing the evidence.

    (An association exists when one event is more likely to occur because another event has taken place. However, although the two events may be associated, one does not necessarily cause the other; the second event can still occur independently of the first. A causal relationship exists when one event [cause] is necessary for a second event [effect] to occur).

    So, the Registries should be cautioned here. We will be looking beyond the Quantitative data; we will be determining cause and effect in our evaluation. They may not put all their energy on just the numbers.

    But most especially, we will be looking at Confounding factors. Elsewhere, it is described that “to rule out that a relationship between two events has been distorted by other, external factors, it is necessary to control for confounding. Confounding factors may actually be the reason we see particular outcomes, which may have nothing to do with what is being measured”. And I agree.

  7. Joseph Peterson

    When it comes to legal issues, I’m utterly ignorant. So I’ll openly ask the lawyers in the audience to tell me what’s what.

    Suppose a company courts investors, wooing them with lofty sales projections. Now suppose that there’s some question of paying to inflate purchase numbers or falsifying records by signing up “buyers” without a any purchase. Potentially, we’re talking about fraud, aren’t we?

    Do those investors have grounds for legal action of any kind? My impression (which could be wrong) is that other industries are regulated in ways that would lead to serious civil or criminal consequences if today’s domain-industry behavior were emulated. And maybe that impression of mine is completely wrong.

    Giving away free .XYZ domains with consent is one thing. Signing people up without consent is another. And doing so on the down low while simultaneously promising paying customers that there will be a million .XYZ registrations during year 1? That sounds like selling a service based on fabricated data. And fabricated data is not a form of marketing that does well in the courts, one would imagine.

    I keep thinking about how this would play out if we were talking about publicly traded stock and secret “ghost buyers”.

    Anybody out there actually know this subject?

  8. Reality

    Spot on – the pro-nTLD crowd are keen to compare the sum of all nTLDs to .COM, which is clearly misleading. When an individual nTLD gets 110 million registrations they can start to crow. Until then they’re only fooling themselves.

    • Philip

      Early days yet with hardly any wider awareness of the new gtlds beyond those with an existing interest & clued up individuals – pretty similar to the early days of .Com.

      It’ll be interesting to see how that major, inevitable shift in awareness will impact the numbers.

      I certainly wouldnt want to be betting against the new gtlds when that happens.

  9. Philip

    As usual, its in the wording as well as the T&Cs. They’re projecting a million in the first year rather then unequivocally promising it, there being no specific constraints on the proportion that are gratis or charged for.

    This is part of an initial promo campaign, not far removed from loss leading initiatives applied to certain brands within a wider product base.

    Some might see it as a canny move to help establish the brand, but hardly an effort to ‘defraud’ & ‘criminal’.

    Business heads tend to be more pragmatic in their assessments.

    • Joseph Peterson

      Wouldn’t it depend on how the numbers are officially reported as well as statements / implications made by the company about what is / isn’t going on?

      I mean, if I get a mutual fund prospectus, it needs to identify the flow of money accurately.

      If .XYZ promotes itself by citing its registration numbers as if they represent buyer interest, knowing full well that numbers have been fabricated, then I think that goes beyond a loss leader. Nothing wrong with giveaways or loss leading.

      But if I pay a stock broker to “buy” stock in my company on behalf of tens of thousands of that broker’s clients … without their consent … and then point to the market demand in order sell more stock?

      Explicitly pointing to non-purchases as if they’re purchases is certainly unethical. In some areas, I assume it’s illegal or at least entails the risk of civil litigation.

      If .XYZ weren’t pointing to the numbers in order to make sales, then there would be nothing wrong with giving away freebies. Lots of companies give away free samples to people who consent to receive them.

      If Daniel Negari has other investors on board, and if he gave them to understand that these numbers were genuine purchases (which is what he implied publicly), then those investors have basically been given a fabricated prospectus. Am I wrong?

      Eventually, I’d love to see domain buyers given the same rights with respect to a registry that stock holders have with respect to a publicly traded company. Registry claims ought to be regulated for the sake of accuracy.

  10. accent

    The big one: “IF” Mr Daniel Negari was acting honorably he would be here explaining. He has not missed many chances to publicize his wares up to now. In my eyes his silence is damning.

Domain Name Wire | Domain Name News
%d bloggers like this: