Patrick,

The report doesn’t totally miss the mark but rather quite accurate. .com isn’t confined to the US but is a generic TLD used all around the world, including in every European country. And almost every German company who wants to look outside Germany will have a .com domain name. Plus many others.

And whether it’s “film” or “movie”, Brits and every other English speaker knows what it is and the two words are very interchangeable.

.com domains are popular throughout the world and would be considered “must have” for anyone wanting to expand their business internationally possibly with the exception of the Spanish speaking countries who often treat .es similar to a .com name. A .com domain, from people I talk to, is also much better for search engine optimisation. There is a website that, last time I looked was out of date by a few years, that outlines how many .com domain names have been registered in each country. I’ve forgotten the address, but needless to say pretty much every country has a significant number of .com names registered, but it varies markedly for various reasons.

David

Patrick is right.
The Gartner report misses the importance of ccTLDs completely (I don’t regard Gartner or its analysts as being expert on the domain industry). The breakdown of .com vs ccTLD is interesting because in many countries, the ccTLD registrations overtake the number of .com (or gTLD) registrations as the markets evolve. There is a tipping point at which a country’s ccTLD will begin to exceed the number of gTLD domains registered in a country. After that, .com will keep pace for a while and then the ccTLD becomes dominant. In terms of new launches, the ccTLD will be the dominant extension and the .com becomes secondary. The figures below are from a few days ago. The gTLD registration counts are from webhosting.info and are not statistically reliable. However they do offer a rough guide to the number of gTLD registrations.

Country - TLD/gTLD - ccTLD
.uk 3555629 - 6941940
.de 5745656 - 12133637
.fr 2273012 - 1172565
.jp 1291085 - 1033412
.mx 145469 - 258565
.ca 3130792 - 1047985
.kr 840079 - 932001
.se 288735 - 744812
.dk 339681 - 927013
.fi 169865 - 183234
.it 939443 - 1549441
.br 300174 - 1387453
.au 2380417 - 1168506
co.za 67066 - 410813
.il 80456 - 129494
.es 1165454 - 970580 (May/08)
.nl 885122 - 2997668
.cn 2854920 - 11821635

A more reliable set of figures covering .ie (Ireland) show the tipping point theory of gTLD to ccTLD in action:

Month - .ie - CNOBI - .com

01/Jan/2008 - 91352 - 95817 - 76250
01/Feb/2008 - 93634 - 98394 - 78336
01/Mar/2008 - 96201 - 100545 - 80031
01/Apr/2008 - 98241 - 102545 - 81604
01/May/2008 - 101156 - 105095 - 83625
01/Jun/2008 - 103156 - 107208 - 85292
01/Jul/2008 - 105167 - 108813 - 86583

As for the search engine optimisation, the .com almost always in second place to the ccTLD as search engines are getting more and more regionalised.

In terms of domain statistics, the databases here cover .com/net/org/biz/info/mobi/ie/eu/co.uk and provide statistical reports for the industry each month. The coverage on the gTLDs is 100% and the databases track approximately 2.6 Million .eu domains and 4.1 Million .co.uk domains each month. The market is definitely changing towards the ccTLDs. All of the recently launched gTLDs are third choice extensions for new businesses and registrants. The ccTLDs and .com dominate followed by .net and .org. For Ireland, the official Eurid figures claim that 57K or so .eu domains are Irish owned. These figures are untrustworthy. A number of cyberwarehousing operations such as Marchex and Dotster have used Irish front companies to warehouse .eu domains. The number of detected Irish hosted .eu domains is only around 6K. The second most important ccTLD in Ireland is actually .co.uk with approximately 8K Irish registered domains. What the Gartner report misses is the strongly interconnected nature of ccTLDs in that businesses wishing to do business in markets outside their own will register a ccTLD rather than rely on the anonymity of a .com - that is one of the biggest changes of the last five years.

Regards…jmcc

ICANN’s decision to allow any entity to register a gTLD has already wasted the time of people and businesses all over the world and no end is in sight.

Successful companies, whether established or startups, relay new messages to their stakeholders by presenting a solution to a problem and/or by presenting a motivational springboard story that ushers listeners into, say, the beautiful world of expanded gTLDs.

Instead, ICANN announced a solution that is looking for a problem. This resulted in discussion and speculation.  Is expansion supposed to solve trademark issues, brandability tangles, the conundrum of how many gTLDs will be successful? Or has ICANN decided to take an unexpected leap into the for-profit arena? All this wasted time and energy would have been better channeled into finding the best solution, or at least an acceptable solution, to the mysterious problem that ICANN is trying to address.

This announcement gives us a clear idea as to how ICANN works, and to many it is a sad case of déjà vu. But why waste stakeholders’ precious time? There are a few plausible explanations: communication ignorance, a desire to distract stakeholders from more serious problems that ICANN is pushing under the rug, or a simple yen for dictatorship. Rulings from on high have their place when certain coordination problems need to be tackled, but it is not clear how such problems would arise in deciding on gTLD expansion.

Patrick and John are wrong.

Sorry guys, but what percentage of large companies/organisations/anything around the world, or any company that operates outside of their home country, would not use a .com or other gTLD name? The answer will be a small minority. And the minority is smaller as the company gets bigger.

And of those who don’t use a .com or other gTLD name, what percentage would love a .com name? The answer is almost all, and the majority will get larger as the company gets bigger.

You can use whatever statistics you like, but the fact is .com is the preference for many many websites. And a .com name is definitely not worthless in Germany. Far from it. But yes, a .de domain will be more popular in Germany than any other country. I’m not sure what John’s stats try and say. All they do is back up what on this point Gartner was saying.

For a simple TLD branding model, click here. For an empirical study of gTLD hierarchies (relatively old), click here.

No David,
That’s a strawman argument. Most companies operate in their own locality rather than globally. When they do begin to operate outside their national boundaries, they tend to use the local ccTLD of their target markets. This is due to the benefits of their company appearing in the ‘pages from $country’ searches for their target market. Even the large companies who own their .com domains do this. While having their .com is nice, sales are nicer still. That is the commercial reality. As for .com being better for SEO, that’s rubbish. It is a nightmare when targeting multiple markets since search engines use geolocation techniques, often based on associating the IP of the webserver with a particular country, to determine which country a website targets. With a .com, search engine optimisation generally involves some type of country specific directory or redirection. Owning the ccTLD is a lot simpler it is far more effective for targeting that particular market.

In Europe, this ccTLD effect is far more visible as companies do cover adjacent countries. In Ireland, the second most popular ccTLD is .uk rather than .eu (the official Eurid figures on Irish .eu ownership are rubbish)  Then you’ve got the Belgian/France/Netherlands/Luxembourg markets overlap.

I do use statistics. [1] The .com is a global TLD. Now taken in global terms, it is very popular and the sheer magnitude of the number of .com sites makes any ccTLD look small. However when you break down .com usage on a country by country basis, the importance of the ccTLD emerges. There is a tipping point where the ccTLD becomes more important than .com. For Germany, that point was reached long ago and .de is perhaps three times as important in that market as .com. In the UK, .uk is the dominant TLD.

In my opinion, Gartner hasn’t a clue about domain statistics. Its “report” is a classic US-centric press release. As for the ‘francais’ example, didn’t ICANN mention that new TLDs could not interfere with existing TLDs, in this particular case .fr ccTLD?

Regards…jmcc
[1] I compile stats on .com/net/org/biz/info/mobi/ie/eu/.co.uk registrations (new/deleted/transferred) every month. The databases here are tracking approximately 2.7M .eu and 4.1M .co.uk domains. The domain counts are grouped on hoster DNSes which are then grouped on a country by country basis and are used to prepare domain/hosting industry reports. This might give me a somewhat better insight into domain statistics than Gartner et al.

David,
There is a difference between domain speculation and domain development. Most of the .com domains that sell do so for purely speculative purposes. This is why the number of times that phrase appears in Google or Yahoo is used to market these domains. They often end up being parked with PPC advertising. And these .com domains often depend greatly on direct navigation traffic (where people will type the domain into the browser address bar expecting that there is such a website). The .com has become the default US extension. That and direct navigation are the reasons why these .com domains have such high prices. Most of the high value sales in .com are generic or generic terms rather than company names.

Your idea that .com is the most desirable TLD around the world is based, it seems, on ignoring the importance of the ccTLDs. When .com is broken down on a country by country basis, the statistics show otherwise. However when viewed as a single TLD, it is the largest TLD. Where it is the most desirable, these markets are often early market situations. In a mature domain market, the ccTLD beats .com and the gTLDs.

Regards…jmcc

Most of the domains registered in any new TLDs are speculative. They are registered by people who hope to sell them rapidly. When they cannot sell them, they end up parking them. If they don’t generate enough revenue to cover renewal, most of these speculative domains will be dropped at the anniversary of the TLD’s landrush.

Many of the new registrations in .eu and .asia followed .com speculation rules targeting generic names and phrases. There were also brand protection registrations. Most of the websites in a new TLD will be parked with PPC advertising. The website development percentage will be below 20% and often closer to 10% of the total number of domains registered. As for checking .eu usage, I have checked it.

I surveyed approximately 2M resolving .eu domains in January 2008. Of these domains, 1723638 had websites. Most hosting packages will automatically set up a website. Some hosters will point this to a PPC page. In .eu’s case, I think that around 14% of the sites were PPC/parked with around 13.37% being actively developed/unique. I think that most of the redirects were to ccTLD sites rather than .com sites. These redirects accounted for approximately 16.68%.

As for .asia, I haven’t run any web surveys yet. However there is a speculative trend in the domain registrations. I think that the number of active .asia domains was around 200K at the start of the month. The last TLD survey that I ran was on .mobi in July 2008. The preliminary results were:

Number of .mobi domains in survey: 927369
Number of .mobi websites responding: 683584

type | sites |
+————-+————-+
| F | 3754 | Forbidden/payment required
| X | 4042 | Porn
| S | 5434 | For Sale
| N | 6329 | Networks of sites (duplicate content)
| U | 6352 | Unavailable
| B | 11274 | Brand protection registrations
| D | 15295 | Dead sites/domains
| R | 58933 | Redirects (not classified yet)
| H | 59272 | Holding Pages (coming soon etc)
| A | 69841 | Active (HTTP 200)
| P | 443068 | Parked/PPC

The reality of domain registration is that most domains, especially in a generic TLD like .com, will drop without every having being developed or used. Because of its unmanaged nature and low cost, the majority of domain registrations will be speculative. Due to its maturity and global nature, a high percentage of .com domains will be parked with PPC (possibly approaching the percentage of domains with developed site). The .com has had many years of development. With these new TLDs, a high percentage of domains go straight to PPC without every being developed.

The majority of the parked sites in the .mobi survey were on Godaddy. The .mobi is coming up to its landrush anniversary and many of these highly speculative domains will drop. This Junk Dump effect was also visible with .eu when approximately 250K domains dropped around the anniversary of its landrush fiasco.

Regards…jmcc

John,

How many of your .eu websites, especially those for larger companies that cross borders, did and did not redirect to another domain name such as .com?

You miss the point. A larger company whose business crosses borders will want a .com name. And when people type in a domain name, as many do, they will type in their own country’s ccTLD or .com. It will be rare for a Brit to type in a .de domain name.

And VeriSign announced there are about 87.3 .com and .net domain names. No matter how you look at it, the demand is much higher for .com than any other gTLD, no matter how the domain is used. The demand for domains that are just used for advertising just reflects there is a demand. Even if they are speculative, there is a reason they choose .com over other TLDs.

I agree with John McCormac’s point that, this Gartner report is a classic US-centric press and therefore the recommendations made mostly fit American business interests.  I believe that, .com.ccTLD would be the top priority for companies regardless the emerge of new gTLDs.

Let me bring .tw as an example.  The total amount of .tw registration is 387,407 (though quite small, still referable), and if we look at the second level of the .tw names, you will find that, .com.tw is still dominant.

If we ignore what Gartner suggested precisely on the terms of gTLD or ccTLD, perhaps .com.ccTLD will still be the priority for business for their web name registration.

We can also observe the second level names usage to further predict the possible not-so happy development of the non-conventional gTLDs (non-conventional refers to those other than .com, .gov, .org)

—data from TWNIC, updated by August 2008——
.com.tw—>  153,189
.net.tw—>  1957
.org.tw—>  10,593
.game.tw—> 265
.ebiz.tw—> 166
.club.tw—> 522
.idv.tw—>  20,249
.gov.tw—>  2,006
.edu.tw—>  473

I’m still trying to work out what the number of registrations in Taiwan, Ireland, or any other ccTLD has to do with the debate over whether ccTLDs are more popular than gTLDs. It’s pointless to use individual ccTLD statistics.

What would be more interesting, and go some way to proving the point one way or another, is what proportion of .com names are for businesses and individuals with US addresses and non-US addresses. Then compare this to ccTLDs. What you will undoubtedly find is .com registrations have a much more global reach than ANY ccTLD.

Almost nobody outside Ireland or Taiwan could give two hoots, that is they don’t care, about .ie or .tw. Internet users either have an interest in their own ccTLD or a gTLD. And .com has proven to be the TLD that is most coveted whether you want to monetise the site, develop a website for your business or interests and/or buy it for possible later sale. Internet users in Ireland will type in a .ie domain name, and people in Taiwan will type in a .tw domain name. But they largely won’t type in a .de domain name.

And if you have a business or interest that has a desire to reach out around the world, for most people the first choice for registration will be a .com name.

It is very simple David,
The stats on .ie or .tw versus .com or the gTLDs in those markets show just how .com is doing in those markets. While it might be nice to take .com as a global TLD, the reality is that ccTLDs are overtaking it in their relative markets. Now as for the claim about .com being the most coveted extension for monetisation, that is largely true when it comes to type-in/direct navigation traffic. However even that is changing as ccTLDs take over their relative markets. You are simply looking at .com as a global market. Our view seems to have a greater granularity in that we see .com as set of markets. The .com has become the default US extension. However as ccTLDs have liberalised over the past few years, they have grown. And they have grown at first in parallel with .com but gradually they reach a tipping point where they overtake .com in their markets.

I think that I did provide a rough ccTLD/gTLD comparison above. Having a global reach is all fine and well. But this is business. Getting the marketing right is essential. The era of a one domain fits all has passed and businesses trying to enter a market will try to register their domain in the local ccTLD. The benefits for this is that they automatically are included in the ‘pages from $country’ search on the main search engines and the sites rank better. Now you can waffle about .com being a global choice but the reality is that the localisation of marketing is happening regardless and this localisation depends on ccTLDs.

And as for those .eu domains that some of these multinationals have managed to register, they are generally used as portals so that the EU users can be diverted to the local ccTLD website. That’s where they are even used at all. The usage in .eu is well below a typical ccTLD due to the .eu landrush fiasco. Perhaps we are arguing about different things. The difference is that we have provided stats and you have not. As an efficient way of using a non-targeted extension, .com is best. But targeted marketing requires just that - targeting.  And ccTLDs are the most efficient way to target a specific country’s market.

Sorry John, but while you have provided stats, I don’t see they prove anything. I could provide stats on the sale prices of domain names that show what people pay big money for and this what shows demand. But then you’ll have an excuse not to believe that anyway.

I do accept that a ccTLD in its own country will be desirable. But what relevance does a .tw domain have for someone in Ireland? Bugger all. But there is some relevance for .com.

Anyway, this debate is like a dog chasing its tail. Your statistics only show that ccTLD registrations are increasing. Some of this is as you say, an interest in ccTLD registrations, and some of it will be because the .com name is not available.

I do know from people registrars that they recommend a .com name over any other if you can get it. And these people know more than you or I. And they also have discussions with the search engine companies about this to exchange knowledge.

Yes domain names are global, yes marketing is important, but if someone can get their .com name and their business or whatever interest go outside their country, they will. As I said, ccTLD domain names are great for within your country. But outside it has little relevance. And I haven’t seen any statistics that show anything to disprove this. In fact, if anything, all your comments do is prove this.

There is also an element of foreign ownership in the growth of ccTLDs David,
This is a subtle trend that is only apparent when you cross-reference the webserver and mailserver IPs. As for the relevance of .tw to people in Ireland, most of the BIOS downloads for motherboards seem to be from .tw sites. Businesses targeting specific national markets now tend to use that market’s ccTLDs. The .com is nice to have but the ccTLDs tend to be far more important in adjacent markets. My stats show that ccTLD registrations are increasing and overtaking gTLDs in some cases. This pattern of ccTLD dominance is more apparent in mature markets. And as other markets mature, their ccTLDs will eventually become more important than .com - but only in those markets. This seems to be two parallel arguments - one about the global nature of .com and one about how .com is really a set of national markets. The reason that some search engines like .com is simply because it is easier to detect new .com websites due to zonefile access to many ccTLDs not being available. But search engine development and the importance of ccTLDs is a completely different argument.

Patrick, I also know that plenty of companies defensively register their ccTLD names for their brands and name. It goes both ways. And while you give some examples, there are others such as IBM that when you type in ibm.ccTLD, it redirects to the country section of the .com site.

The reason why we debate on whether ccTLDs are more popular than gTLDs is because Gartner’s recommendation to encourage business invest firstly on the “.com”.  My point is that, Gartner’s recommendation is not that appropriate for all the businesses in the world and this argument is less relevant to the fact that .com registration is much much more than ANY ccTLD registration (BTW, I totally agree this statement.)

Most countries in the world, the number of small and medium enterprises (SMEs) occupy 95% or even higher portion of the overall business amount.  e.g. Australia (95%), Germany (99.7%), Japan(99.7%), Korea (99.8%), Taiwan (97.8%), UK (99.8%)

Just to show the significance of the SMEs, in most countries, SME contribute tremendous economic value to GDPs. 

In the case when those SMEs try to get web presence, I believe that, the ccTLD will be their first priority than registering a gTLD.  Below you will also find another interesting survey result made by SIDN:

*  .NL still preferred by 87% of businesses.
*  Businesses often start with .nl before registering Gtld’s.

I have a feeling there are two or even three debates going on here. The original point made by Gartner was they expected “‘classic’ TLDs, particularly ‘.com,’ to continue to overwhelmingly dominate the domain name marketplace.”

So what SMEs do or don’t do is largely irrelevant. Of course they will largely register ccTLD names. Mostly they are silly not too.

The discussion has been about, or should have been about, will ccTLDs usurp, or have they usurped, .com as the most coveted/desired TLD. My argument, and Gartner’s, is .com is still the most coveted TLD. In fact, Gartner for better or worse only looked at gTLDs in their analysis and didn’t even discuss ccTLDs.

The future of .com is probably worthy of a debate of its own. Gartner’s “analysis” has all the credibility of a press release when ccTLDs are included in the equation. It seems to be trying to reassure US subscribers that .com will still be king.

Most of these new TLDs will end up being niche extensions. Gartner’s analysis is based purely on seeing the total figure for .com and ignoring the ccTLDs. It is like looking at an ocean and not even realising that there are many forms of life in it.

As I was travelling on the train today through Germany I drank my Bionade, the label entirely in Germany. But was TLD do they promote? Well, .com of course. And when one types in bionade.de, as soon as you click on a link, it goes to the .com address. Then I pick up my Vittel water. Well, what do you know, a label entirely in French and maybe Dutch. What TLD? Why, .com of course. Then I read The Economist, a truly global newspaper. And what TLD is on the cover? Oh no, not another .com! Now I’m sitting in the British Airways lounge. I try to use the .uk version of their website, but, lo and behold, it redirects to .com.

Sure, companies who have local versions of their products will often use the local ccTLD, but for the majority of businesses around the world who do business across borders, .com will always remain their number one domain. Besides, most people know of .com. But awareness of ccTLDs outside the home country will often be low. gTLDs AND ccTLDs will both be important for internet users. But why a few people here can’t accept that .com, regardless of Gartner’s knowledge of the area, will remain for the indefinite future the most desired TLD is beyond me.

Because the future is never definite.

True John. The aliens may really be out there waiting to come and get us. And those scheming buggers they are going to come just as .com is about to lose its role, in about the year 2378…  Or the rapture might finally come/happen/whatever it’s supposed to do.

Slightly more likely is the internet as we know it will have evolved into something different, or something else will completely replace it, not even yet developed or thought of, before this happens…