DomainsBot, which powers the name suggestion feature on most major registrar storefronts, has unveiled a significant update designed to make selling new gTLD domains easier.
The company reckons its new technology will soon be promoted from a follow-up sales tool, rolled out if a customer’s first choice of domain is not available, to “replacing the availability check” entirely.
“The idea is to be at the heart of the process of promoting new gTLDs,” CEO Emiliano Pasqualetti told DI.
The idea is pretty straightforward: a customer types a word into a search box, the service suggests available domain names with conceptually similar TLDs.
While it may not be perfect today, it was pretty good at finding appropriate TLDs for the keywords I tested.
And Pasqualetti said that under the hood is a machine learning engine that will make its suggestions increasingly more relevant as new gTLD domains start to go on sale.
“It tries to predict which TLD we need to show to each individual using a combination of their query, their IP address and as much history as we can legally collect in partnership with registrars,” Pasqualetti said.
If, for example, customers based in London show a tendency to buy lots of .london domains but hardly ever .rome, Londoners will start to see .london feature prominently on their registrar’s home page.
“We learn from each registrar what people search for and what people end up buying,” he said.
Some registrars may start using the software in their pre-registration portals, increasing relevance before anything actually goes on sale, he said.
My feeling is that this technology could play a big role in which new gTLDs live or die, depending on how it is implemented and by which registrars.
Today, DomainsBot powers the suggestion engine for the likes of Go Daddy, eNom, Tucows and Moniker. Pasqualetti reckons about 10% of all the domains being sold are sold via its suggestions.
Judging by today’s press release, registrars are already starting to implement the new API. Melbourne IT, Tucows and eNom are all quoted, but Pasqualetti declined to specify precisely how they will use the service.
It’s been widely speculated that Go Daddy plans to deploy an automated “pay for placement” system — think AdSense for domains — to determine which TLDs get prominence on its storefront.
Pasqualetti said that’s the complete opposite of what DomainsBot is offering.
“We’re relevance for placement,” he said. “We want to give every TLD a chance to thrive, as long as they’re relevant for the end user.”
According to Pasqualetti (and most other people I’ve been talking to recently) there are a lot of new gTLD applicants still struggling to figure out how to market their TLDs via registrars.
There are about 550 “commercially interesting” applied-for gTLD strings in the DomainsBot system right now, he said. New gTLD applicants may want to make sure they’re one of them.
Next week, the company will reveal more details about how it plans to work with new gTLD registries specifically.
A lawyer apparently representing a rival new gTLD applicant has questioned ICANN’s background screening processes after Demand Media managed to get a pass despite its history of cybersquatting.
Jeffrey Stoler, now with the law firm Holland & Knight, last July said ICANN should ban Demand Media and its partner Donuts from applying for new gTLDs under the rules of the program.
This month, he’s written to ICANN, the GAC and the US government to express “alarm” that both companies have managed to pass their background checks. Stoler wrote:
This alarm arises from the overwhelming evidence, as referenced below, that: (a) Donuts is a “front” for Demand Media, Inc. (“Demand Media”), and (b) Demand Media’s status as precisely the kind of proven cybersquatter that ICANN’s rules were designed to weed-out of the gTLD application process.
How ICANN’s background screening panel could — in the teeth of that evidence — approve the continued participation of Donuts in the new gTLD program (the “Donuts Decision”) requires justification. This letter formally requests that ICANN, pursuant to its obligations of accountability and transparency, provide an explanation of how, and on what basis, the Donuts Decision was made.
Both Donuts and Demand Media responded with anger and disdain.
CEO Paul Stahura told ICANN that Donuts has discovered that Stoler, who has still not disclosed which client he’s representing in this matter, is actually on the payroll of a rival.
Donuts suspected his client was a competing applicant seeking to gain commercial advantage, and we have since confirmed this in fact is the case.
Not only do the letters intentionally misrepresent facts, they are a preposterous, extra-procedural tactic that is a regrettable waste of time and community resources.
David Panos, director of Demand’s applying subsidiary, United TLD Holdco, was similarly dismissive:
Clearly, Mr. Stoler’s client has a substantial commercial interest in the new gTLD program and is seeking to eliminate its competition by mischaracterizing the relationships of other competing applicants and by restating factually inaccurate statements
What’s notable from both the Stahura and Panos letters is that neither company actually addresses Stoler’s allegations directly, resorting instead to mainly fudging and ad hominem arguments.
Stoler probably is seeking a competitive advantage for his mystery client, and his claims about Donuts being a “front” for Demand do come across as a bit of a stretch even for a lawyer, but that doesn’t mean that all of his arguments are wrong.
ICANN’s Applicant Guidebook for the new gTLD program is pretty clear: if you’ve had more than three adverse UDRP decisions, with at least one in the last four years, you’re “automatically disqualified” from the program.
Demand Media, as Stoler alleges and the public record supports, has lost about three dozen UDRP cases through subsidiaries such as Demand Domains, the most recent of which was in 2011.
So how did Demand pass its ICANN background screening?
The Guidebook does say “exceptional circumstances” are enough to get an applicant off the hook, but it’s hard to see how that would apply to Demand’s over 30 UDRP losses.
And Demand doesn’t want to talk about it.
None of its responses to ICANN that have been published to date even attempt to say why Stoler is wrong, and the company declined to comment when we asked for clarification today.
Donuts, which is using Demand as its back-end registry and has given the company the right to acquire interests in over 100 of its new gTLDs (should they be approved) didn’t want to comment either.
Which, some might say, plays right into Stoler’s hands.
If there’s a simple, straightforward explanation for why the background screening rules apparently didn’t apply to Demand Media, is it unreasonable to ask what that explanation is?
Are the synergies between domain name registrars and content farms not all they were cracked up to be?
That’s a question emerging from last night’s news that Demand Media is planning to spin off its domains business, which includes number-two registrar eNom, into a separate public company.
The company, reporting its fourth-quarter earnings, said that it’s looking into the possibility of breaking up content and domains into two separately-owned businesses.
The new domain company would comprise eNom (wholesale) and Name.com (retail) on the registrar side, any new gTLDs Demand manages to win, the formative registry back-end business, and its stake in NameJet.
“We believe that a separation of the two independent companies will better position each business to pursue their increasingly diverse strategic priorities and opportunities,” CEO Richard Rosenblatt told analysts.
The spin-off would have revenue of over $150 million a year and margins of almost 20%, chief financial officer Mel Tang added.
Rosenblatt said new gTLDs will be “transformative” for the domain industry, saying that Google, Amazon and dot-brands will help grow consumer awareness of the world beyond .com.
Demand has filed 26 new gTLD applications of its own, and has 50-50 rights to 107 more with Donuts.
Previously, Demand has stated in regulatory filings that it used data gathered from domain name lookups to create ideas for the content farm side of its business. It’s not clear if that would continue after a split.
Directi has signed up more registrars for its launch of the .pw top-level domain than .xxx managed a year ago, crediting pricing and “operating in new gTLD mode” for its progress.
Sandeep Ramchandani, business head of .pw at the company, said that over 90 registrars are currently accredited. That’s compared to the about 80 that ICM Registry launched .xxx with in December 2011.
The ccTLD for the tiny nation of Palau (pop. 20,956), .pw isn’t what you’d call an intrinsically exciting string, despite Directi’s attempt to rebrand it as “Professional Web”, making its relatively strong launch channel a bit of a head-scratcher.
The fact that Directi also runs registrar-in-a-box provider LogicBoxes has helped it add some registrars, no doubt.
But Ramchandani reckons a combination of low pricing, open registration policies, a focus on developing markets, and attractive registrar incentives, are helping the TLD gain channel traction.
“There’s going to be a massive shift in power from registries to registrars,” he said. “We’re basically operating .pw in the new gTLD mode.”
I had assumed that many registrars might have wanted to plug in to .pw in order to ease the need for integration work with Directi’s registry if/when the company launches the 30-odd gTLD it has applied for, but Ramchandani pointed out that this is not the case.
The company is using CentralNic as the back-end for .pw, but it’s signed up to use ARI Registry Services for its gTLDs.
“This is not something related to our gTLD business, because we’re working off a completely different platform,” Ramchandani said.
But with the imminent launch of new gTLDs, there’s recently been an increased industry focus on how registrars should be paying their registry fees. Typically, they prepay each TLD registry before they sell domains to registrants, tying up capital that they could be using for other purposes.
That model may work in a world of 1,700 registrars and 18 gTLDs, but it will become increasingly cumbersome and uneconomical for registrars as the number of TLDs approaches 1,000.
Some say it will make more sense for registries to scrap the prepay model, if they want to attract more registrars.
While Directi has stopped short of offering blanket post-pay to its registrars, Ramchandani said it will offer the model in some cases (unlike ICANN-contracted gTLDs, .pw has leeway to treat registrars differently).
“We will be asking for prepayment, but if a registrar is doing significant volume… If they feel they’re blocking a lot of capital and require a more convenient and flexible model we could offer post-payment,” he said. “But it’s not something we can offer every registrar.”
It’s also offering what it says is a more attractive way of handling promotional pricing.
Typically, if a gTLD registry runs a pricing promotion today it will rebate its registrars the difference between the promo price and the regular fee monthly or quarterly based on their sales volumes.
Ramchandani said the .pw model is more attractive to registrars: “With our programs, we will charge the discounted amount up-front, as opposed to charging the full amount and rebating later.”
“We will be coming up with some very, very aggressive promotions that will bring down the first-year registration cost a lot,” he added.
In contrast to SX Registry’s .sx (for the new nation of Sint Maarten), which is launching with prices of around $50 a year, Directi is selling .pw domains with a registry fee “sub-.com” before discounts.
Exact registry pricing for .pw domains has not been publicly disclosed, and Ramchandani declined to give out that information, but given current .com prices we can estimate a ceiling of about $7.85.
Directi reckons the low prices will drive volumes in developing markets, such as its native India.
The .pw launch is currently in the last few days of sunrise — which like so many other recent sunrise periods has been extended to cope with last-minute filings — during which trademark holders can defensively register their brands as domain names for a higher fee.
Not ever registrar is participating in sunrise; eNom, for example, which is .pw’s biggest registrar signing to date, does not plan to get involved until landrush.
Directi has something interesting planned for landrush, which begins next Monday, too.
According to Ramchandani, the registry will release and promote a list of unreserved “premium” domains that are available for registration during the landrush period.
This is slightly different to the standard registry practice of holding back premiums for auction. The names .pw will promote will not be “reserved”, they’ll just be examples of decent strings picked out of the available pool.
Each could technically be sold for the basic landrush fee if only one registrant attempts to register them. In practice, due to the promotion, there’s a higher likelihood of the domains going to auction, however.
“We’re making available a much larger set of premium names during landrush,” Ramchandani said. “We think it will help raise awareness to say: ‘Hey, these are the top picks.'”
“Since those names are generally reserved by the registry, we think it’s important to say: these are available,” he added.
The full list of registrars participating in .pw’s launch is available here.
Directi is promoting those registrars that have committed some marketing resources to the TLD, such as by creating dedicated landing pages.
It has not yet signed up Go Daddy, which is typically responsible for a quarter or more of all sales in gTLDs it sells, but Ramchandani said he expects more “top five” registrars to follow eNom in supporting .pw soon.
Demand Media executive vice president Taryn Naidu said newly acquired registrar Name.com plans to carry as many TLDs as possible, but urged new gTLD applicants to start distribution talks with registrars as soon as possible.
“It’s going to be challenging to offer all of them,” Naidu told DI today. “We’re asking registries to come talk to Name.com early and often to make sure they get the shelf space.”
“They have to come with a plan, and make sure they’re ready to go to market,” he said.
Demand Media announced the acquisition of Name.com earlier today. The deal, for an undisclosed amount, will see the 30-strong Denver, Colorado-based company join number two registrar eNom in the Demand stable.
Name.com is almost 10 years old and has almost 1.5 million domains under management, the majority of them in gTLDs. eNom has over 12 million domains spread across scores of registrar accreditations.
Naidu said that the forthcoming new gTLD market was a major reason for the deal.
While eNom is primarily a channel player, Name.com is all about the customer-facing retail side of the registrar business.
Owning Name.com could give Demand Media a faster way to market the dozens of new gTLDs that it has itself applied for, as well as the 300 it has partnered with uber-applicant Donuts on.
Naidu declined to comment on details of the Donuts relationship, but I’d be quite surprised if a commitment to carry its TLDs is not part of the deal.
He also said he’s not too worried about alienating eNom’s existing reseller channel, pointing out that main retail rivals such as Go Daddy and Tucows also have extensive reseller networks.
“In many regards having access to a retail player like this will help us serve our resellers by better understanding their needs,” he said.