Donuts has responded to last Friday’s cybersecurity consulting firm Internet Identity (IID) report predicting “the eventual demise of any number of new generic top-level domains (gTLDs) and the websites populated within them—primarily due to financial underperformance”.
Donuts has responded strongly to debunk the report saying as the largest new gTLD operator they will not be closing down any new gTLD strings.
Here is the statement Donuts gave to thedomains.com exclusively about the “report”
“”One has to question the veracity of such a report, given, first, what we know about contingencies for just such a potential occurrence and, second, the report is woefully short on data and facts.
It’s important to point out that even if a gTLD fails financially, ICANN has procedures in place to ensure it doesn’t go suddenly dark (e.g., the emergency back-end system, and cash / letters of credit on deposit with ICANN).
There’s little to be concerned about on this front.
In fact, ICANN has backup systems in place that are more extensive than what exists for registrars that go dark (which happens fairly frequently), and even those instances leave no damage to end users.
Donuts will not sunset or shutter any of its gTLDs, and our company is skeptical of speculation about this kind of “market.” From our perspective, a portfolio approach brings stability to the equation, thanks to efficiencies and lower costs.
Donuts is highly profitable even though we administer some gTLDs with relatively fewer names. Specific gTLDs, such as .BANK for example, can command higher prices than can other generics, and perhaps are therefore more profitable per name than gTLDs with larger volumes. We speculate that some gTLDs with high volume are losing money, while some with lower volumes are very profitable. At best, there is no correlation between volume and profitability, as the report supposes.
Regardless, the market is seeing accelerating growth in new gTLD registrations, which are now surging well past 10 million (the document quotes a six-month-old statistic of five million). We’re seeing impressive growth in new gTLD registration, with the trend lines continually improving.”
If things go wrong, an emergency mechanism is in place (has it been tested ?), and icann can take over and find another suitable registry operator.
The failing TLD has to maintained for three years. But nothing is guaranteed beyond that point.
It is obvious that not all TLDs are viable, some, possibly many, will have to go under and be retired, more or less smoothly.
Phil Buckingham says
Kate , there is an emergency mechanism in place , but never has been tested . The mechanism is a Continued Operations Instrument( COI) , a complex calc ,based on the five critical functions that would need be funded/ costed ( based on registration volumes) when an Emergency Backend Operator ( EBERO) is appointed . ICANN has contracted with three( to cover Europe , USA / Canada, Far East). The COI calc is linked to a Letter of Credit.
The question is when is a TLD a failing TLD ,who decides / how / when is the LOC trigggered . Based on a realistic ,median ,costed projected break even of 10000 year 1 registrations our calcs show 65% so far which will not reach this B/E. They just cant go “dark” , will still need to resolve. Who would be stupid enough to buy a ” failing “TLD in total, so only the solution it seems is a sunset ( as Frank describes below). This is a huge problem for ICANN, particularly as there previous total market projections are way off base/ budget and still are.
Phil Buckingham says
Apologises – written at a very late hour.
It should read “this is a huge problem for ICANN, particularly as their original total market projections are considerably below budget.ICANN wasnt expecting so many TLDs to be in financial trouble so quickly so potentially triggering the COI/ LOC and the appointment of an EBERO.
Sounds like they did not plan at all. Does anyone know what would happen if a new gtld registry went bankrupt? Presumably the bankruptcy court would not permit the customers of that business go out the door with no compensation to the bankrupt, meaning its secured creditors. The customers are valuable.
So these emergency back end operators maybe can donate their services to the bankrupt estate, but they do not have any rights to the customers of it. Heck, I’d bid for the name, address, phone number and email address of anyone stupid and solvent enough to by lots of theses new gtlds. Do emergency backend operators expect to pick up those customers for nothing?
I think that’s the crux of the matter.
Some failing extensions will not find another operator if they are not viable. Then they will be shut down.
EBERO is not about permanent continuity, it is more about switching off the lights gracefully before people exit the building.
Most of the quotes I see for cost of running a registry include things like salaries for CEO. Given the technology is fairly standard in operating these things – afilias, neustar… example could probably run these things at little cost (not sure if Donuts/Uniregistry runs their own or outsources).
I would think in 3 years you could probably manage a TLD on an existing platform in maintenance mode for less than a few K a year. It’s not like each TLD needs another CEO
That is their stance today. We will see how the market shakes out in a few more years.
Donuts is stating here that they have TLDs that are not profitable. How long will they let these losses pile up? $1 million in losses? $10 million? $25 million? At some point it just does not make sense to keep losing money on a losing TLD.
David Thornton says
kd, I don’t read it as donuts stating that THEY have gTLDs that are unprofitable. “We speculate that some gTLDs with high volume are losing money, while some with lower volumes are very profitable. At best, there is no correlation between volume and profitability, as the report supposes.” I read that as their speculation about gTLDs run by their competitors otherwise why would they need to speculate about it? They’d know!
There may of course be a number of donuts gTLDs extensions that, when taken in complete isolation, would be loss making on their own. Donuts however point out that they take a portfolio approach to this.
Is there any thought that in time ICANN may reduce fees to those holding more than a certain number of extensions on a sliding scale or find some other variable method for deciding on the cost of each individual gTLD extension?
I think it’s naive to assume you have 200 endings and none of them will ever go dark. I’m not saying you’d see a blackout, but sunset is a healthy thing.. It’s “renewal” . I could see buying out lucky registrants in a sparsely populated namespace, or offering a premium valuable name of better quality to registrants in underused spaces.. This is a new world. In 10 years time there could be 10,000 or more GTLDs.. in 15 years some of those 10,000 just added, might be wound down in an orderly fashion. It’s healthy to think about and cathartic. I won’t be certain Donuts will wind down a string until they deny it a third time. This is #2
paul stahura says
Frank, consider this #3, #4, and #infinity. I don’t know about UniRegistry, but Donuts is profitable. We think of all the TLDs as one big registry. Its profitable, so all our TLDs are profitable, but that is beside the point. We’d no more shut down one of our TLDs than you would shut down 100 “unprofitable” second-level names in .link. We are currently buying TLDs.
A registry that begins shutting down TLDs runs the risk of undermining their other TLDs resulting in less sales overall.
The key is having a good selection of endings that are likely to cross promote the other extensions.
In the case of Donuts, any NIC or WHOIS address takes you to donut.
The value isn’t just in the sale of a domain, but in the use by end users as well as expanded awareness of across the board that options beyond .com exist.
A handful of extensions that underperform is not an issue, a majority of extensions would be cause for concern if it is dragging the registry down.
Like any business, a few loss leaders can invigorate over all sales.
keeping them alive a couple more years in hope they may turn a profit makes sense. However no business wants to lose money. Eventually we’ll see what happens after the three year guarantee.
No company in their right mind would say they are going to shut it down. What they say and what they do are two different things.
Davd Wrixon says
If you are going bust the last thing you do is admit you are in trouble, as it then becomes a self-fulfilling prophesy.
It is almost impossible to act ethically in these conditions regardless of your best intentions.
Domain Observer says
As there are so many strings, anything can happen in the future. That is called UNCERTAINTY which is hated by investors and endusers. And that’s what may affects the new GTLD industry as a whole. And that only increases the credibility of COM and CCTLDS, not the new GTLDS.
com is just as uncertain.. It just seems certain, today. What does .com look like with $20 registry fees .co.uk just doubled.. Paul, you’re going to be a public co (or get bought) and you personally don’t own Donuts, your investors do. What’s to say they won’t have a change of heart, or CEO.. I might buy Donuts one day and I’m mercantilist. It has to make money. There’s no crime in shutting anything down.. And discussing it openly is healthy for our young industry. All opinions my own.
I understand Donuts position here. When you are trying to establish dozens of new gTLDs against the biggest franchise in the world that .com is, and you are just about 24 month or so into the race, the very last thing you want to talk about is “shutting down” strings already just because they are unprofitable. It hurts the whole gTLD movement and message as such. Registrants will look at it and say what happened to .xyz (not the real TLD just as an example) can easily happen to .website or .guru. Why invest money in building a brand when they can just shut it down or “sunset” which is nothing more like an euphenism for shutting it down plain vanilla. They will simply remember .com which is around for 30 years and is going nowhere.
That speculation, “the eventual demise of any number of new generic top-level domains (gTLDs) and the websites populated within them—primarily due to financial underperformance,” takes away from IID’s credibility. It makes IID sound like it doesn’t understand the domain name system. Since no extension has been canceled, the demise of new gTLDs is conjecture that doesn’t have any foundation.
When people lose websites through domain theft or closing down a Registrar over nonpayment or a botched server migrations, those loss of brands pose the real economic and security threat.
Is it logical to assume that only the least popular gTLDs might go dark, even in the worst scenarios the more popular gTLDs would probably just get reassigned or bought out. One solution might be for the registry to provide some kind of guaranty to the registrants indicating how the registry is going to compensate and help them if a gTLD goes dark. Another solution might be for the registrants to be given the option to purchase some kind of insurance that would compensate them for any loss or expenses that they might have to occur by switching to another domain. (Just an idea)
Davd Wrixon says
Sounds like one of those religious promises where if you put up with a whole load of shit in this life then you will be rewarded in the after life.
The only thing guaranteed in this life is that people won’t get less dumb.
You won’t be able to change the facts by injecting your ugly bigotry into this discussion. Hiding behind a phony name won’t give you any credibility either.
Trying it again: If you are trying to establish dozens of new gTLDs in a saturated market the last thing you want to hear about, talk about or even read about is a public discussion about “shutting down” strings that are 12-18 month old. Registrants will look at this at think why would I risk money, or waste my time building a brand in a gTLD that can be shut down just like that. They will look at .com which is around for 30 years, with 120 million registered domains that is going nowhere. Call it shut down, call it “sunset” a bloody euphemism for shutting sth down but in the end it means the same.
so I am guessing you are not a big fan of Schilling’s ‘solution, dissolution, growth, decay, ….Transformation!!’ theory?
Internet is a public utility and as such any company that is somehow involved in or connected to the infrastructure,operation, maintenance, and security of the Internet has an obligation to the public as a whole that goes beyond just protecting shareholders interest. When ICANN came up with the contingencies and safeguards for the new gTLDs they didn’t think that they were going to end up with so many mediocre strings at $185k a pop. There has to be more protection for the registrants as the number of the new gTLDs grows and especially if and when ICANN reduces the application fees.
We can’t let an important and valuable public utility such as Internet to be at the mercy of the stupidity of those who think this is only about them making money.
Rod Rasmussen says
This entire spin-up here is just a bit amusing to us here at IID. We made a PREDICTION and did not issue any sort of “report” as has been posited here by Donuts. We love donuts (the kind you eat particularly!) and have nothing but great things to say about them, and nothing in what we said mentioned them at all, or frankly was even considered at the time (sorry Donuts folks – nothing personal). The prediction was purely an opinion and was not backed by any formal economic analysis or tracking, etc. (and we never claimed it was) and was based on observation of events, news reporting, and some sound economic theory. On the theory part, in any new, emerging market: there will be failures (just check the stats on new restaurants!). That is an economic reality and if somehow the domain name registration space proves to be the magical unicorn that doesn’t have that happen, then we’ll be happy to see our PREDICTION proved wrong – yeah, domain name industry! For those of us in the real, pragmatic world though, we expect failures in any new field, and we should all be just fine with that if we believe in free markets. Just be aware of that as the excitement of all the new names gets you thinking about the possibilities.
We fully support consumer choice and experimentation that the new gTLD program represent and hope for everyone’s success. However, it is pure folly to expect that all speculative gTLD’s will be successful. If large portfolio holders like Donuts decide it’s in their overall strategic interests to support money-losing TLD’s in order to support their overall position, that’s a move that could make a LOT of sense economically and bully for them for thinking that strategically. That would make a Donuts TLD less risky than others who haven’t come out with similar statements and should be considered in any potential customer’s decision process. Plus that would follow the example of other successful companies, so great long-term thinking there you folks at Donuts!
However, we didn’t make predictions about specific registries, rather just that overall there will be losers in this process that will eventually go dark since consumer interest won’t support them. That’s OK, and to expected in all “normal” markets. Caveat Emptor, however, and you should make your decision on which new gTLD’s to utilize based on the assumption that at least *some* will eventually fail and be withdrawn. To assume anything else would be an irresponsible business decision. If Donuts and other large players want to make a commitment to keeping all TLD’s remain live (a commitment in their registrant agreement of course) then there you go – you’re betting on the overall company rather than a single TLD and that lowers the risk – GREAT!
As a final thought here – IID has been making non-boring, ostentatious year-end, long-term predictions over the past several years with just a touch of subtle tongue-in-cheek. Our “death by Internet” prediction a few years ago garnered us more than a little hate-mail for example (we stand by it though! Check your insulin pump settings…). Please take these predictions in the light in which they are presented – provocative and thought-provoking, with just a little bit of poking-fun at the establishment.
ICANN may want to implement a rule stating any TLD will not be allowed to sunset or shut down irrespective of how bad it performs. Any TLD should be forced to be live forever. If sunset is possible, how will business buy it? Domain extensions for Business are like our own personal names and the idea about sunset is very dangerous to the overall GTLD civilization.
Some strings will go dark. And we all know how behavioral economics works. Fight or Flight. And most flee and panic when uncertainty obfuscates a direct path to safety and security.
What a year in domaining: Rick Schwartz retires and continues to rule. Michael Berkins sells his big portfolio to GoDaddy, who has turned a profit with the acquisition of the Marchex portfolio, and the Chinese phenomenon of acquiring portable wealth (i.e, short number and letter domains). Some GTLD strings can traction. Others – not so much.
In the comments above, I see people making predictions 5 years out. To paraphrase the great William Goldman, “Nobody knows anything…… Not one person in the entire domain industry knows for a certainty what’s going to work. Every time out it’s a guess and, if you’re lucky, an educated one.”
Those who held their cards and held onto the 3 number and 4 number domains, more power to you. You won.
Jay Westerdal says
It is more likely that a sunset TLD will not accept new registrations. ICANN will transfer the TLD from the EBERO to a new entity with a sunset registry contract that will have a lower mandated yearly cost. The domains in that zone will remain until they are not renewed. The TLD can remain in the root and run for little cost by the new entity in sunset mode.
In the future, if the managing entity wants to accept new registrations, they could request to ICANN to move back to a normal contract and get off the sunset contract.
Are you implying that if a gTLD is put in the sunset mode that all the existing domains will be grandfathered in and can continue to function forever as long as they are renewed or would they eventually get deleted (like after 3 years).
Jay Westerdal says
When the Soviet Union ceased to exist and .RU was delegated, ICANN asked .SU to freeze new registrations. While this is a bad example because there is no ability to enforce this, the concept is valid. There is no reason to delete domains owned by registrants, let them sunset forever (not three years).
Either way the registrants have the right to know what could happen to their domains and for how long a website that they have invested time and money in is going to resolve in case a gTLD is put in the sunset mode, whatever the registry is required to do or is promising to do in order to help the registrants should be binding and should be put in writing.