IT.COM

analysis The Most Used Domain Extensions – From Under the Cisco Umbrella

Spaceship Spaceship
It is important to know how much actual internet traffic there is on a domain extension, or top level domain (TLD), since this is one indicator of the health of the extension. There are various measures of internet website use, with the main ones being Alexa, Cisco Umbrella, Majestic, SimilarWeb, and QuantCast. In this post I take a look at the global popularity of different web extensions as measured by the Cisco Umbrella ranking methodology.

Each website use rating system operates on a different principle. For example, Alexa mainly measures visits to sites by those who have installed the Alexa browser plug-in, while Majestic measures backlinks to sites from different subnets.

The methodology for all four rating systems is covered in this paper. That research paper also introduces TRANCO, a new ranking system that uses results from the other four in order to obtain a ranking that is more resistant to deliberate manipulation.

Cisco Umbrella Rankings
Recently CENTR began using the Cisco Umbrella rankings in their data on domain extensions. The Cisco rankings are based on the number of unique DNS (Domain Name System) queries are made for a domain, but only counting those from different IP Internet Protocol ) addresses.

Since the Cisco Umbrella network handles over 100 billion requests every day, the rankings are based on a large dataset. You can read more about the Cisco Umbrella rankings in their post introducing the system.

The top million sites are updated daily in the Cisco Umbrella ranking. I used the Cisco Umbrella rankings for this post partly because they conveniently provide rankings by extension. The data is freely available, and has been provided since 2016.

It is important to realize that the Cisco Umbrella ranking system indicates which extension, or domain, is more popular than another, but does not provide publicly-released measures of the significance of those differences.

The Top 50 Most Used TLDs
I looked at the Cisco Umbrella rankings for domain extensions. The ranking list provided by Cisco is actually a mix of extensions (TLDs) along with some commonly referenced individual sites, such as many specific Amazon server addresses. I extracted only the TLDs, in compiling the following ordered list. Therefore the numbers will be different from those on the raw Cisco list, but the order is the same.

Here are the relative web traffic rankings for different domain extensions, with higher on the list meaning more traffic.
  1. .com
  2. .net
  3. .org
  4. .io
  5. .co
  6. .tv
  7. .ms
  8. .us
  9. .me
  10. .uk
  11. .co.uk
  12. .fi
  13. .ru
  14. .news
  15. .de
  16. .gov
  17. .st
  18. .cn
  19. .jp
  20. .info
  21. .edu
  22. .ly
  23. .fr
  24. .cloud
  25. .it
  26. .ai
  27. .to
  28. .mobi
  29. .xyz
  30. .link
  31. .pl
  32. .eu
  33. .ca
  34. .nl
  35. .network
  36. .be
  37. .biz
  38. .in
  39. .market
  40. .co.jp
  41. .ua
  42. .im
  43. .vn
  44. .gg
  45. .mx
  46. .pro
  47. .club
  48. .es
  49. .am
  50. .media

Findings and Surprises
Many of the results could have been predicted, but there were also some surprises.
  • As expected, .com is in first place.
  • Most will not be surprised that .net and .org, in that order, take second and third places.
  • Beyond that, predictions are more challenging. The general purpose country code extensions .io, .co and .tv, in that order, are the next TLDs with the most traffic, at least as Cisco Umbrella measure web traffic.
  • The next place was a real surprise to me, with .ms taking 7th place. The .ms domain extension , a country code for the British Overseas Territory Montserrat also finds many other uses. It is used by Microsoft for certain projects, and also sometimes for sites related to the state of Mississippi, in addition to several other international uses.
  • The .us country code did surprisingly well, coming in 8th place, well above some country codes that typically sell for higher prices.
  • It was followed by the well known country extensions .uk, .co.uk, .fi and .ru in that order. Given the size of the country .fi is surprisingly active, but the extension finds use in the financial world as well as its regional meaning.
  • I would not have predicted .news as the highest-ranking new global top level domain name (new gTLD) in 14th place, nor that it would edge out major country codes like .de, .cn and .jp.
  • At 16th place is the U.S. government extension .gov, higher than the education .edu and .info extensions that took 20th and 21st places respectively.
  • The .ly extension finds use in domain hacks, and takes 22nd place, while France .fr is in 23rd place.
  • Another new gTLD comes next, with .cloud in the 24th position.
  • After Italy’s .it, the .ai extension takes 26th place. It has found use of late for artificial intelligence startups, or investors hoping to sell to them,
  • The country code for the island nation of Tongo, .to, now sold from Toronto and promoted both as a domain hack or phrase, and for use by the Toronto community, secured 27th place.
  • The decline of .mobi is apparently over-stated, as the extension, while not selling for the high valuations of the past, is still in the top 30 in terms of use.
  • Two more new gTLDs crack the top 30 with .xyz and .link in positions 29 and 30. The .xyz extension also does well in Alexa ratings.
  • A number of major country codes, including .eu at 32nd, take the next number of places, follow by the new gTLD .market in 35th place. There were numerous high value .market domain sales in the past, but it is not as active in aftermarket sales of late. Clearly from this ranking it is still is actively used.
  • The .biz legacy alternate extension was in 37th place.
  • The country code extension for India, .in, is in 38th place.
  • The .gg country code, now used for global gaming, picked up 44th place. This was closely followed by Mexico’s .mx in 45th place.
  • The legacy alternative extension .pro took 46th place.
  • Two other new gTLDs, .club and .media took 47th and 50th places respectively. The Alexa rankings also show .club as a relatively heavily used extension.

Reflections
While ranking may hide big differences in actual use, I was surprised that a number of new gTLDs ranked as highly as they did. Eight new gTLDs placed in the top 50, beating out many major country code extensions.

The ranking of .news is particularly impressive, outpacing .gov, .info and .edu, as well as such major country extensions as .de, .cn, .jp and .fr.

The general use country codes .io, .co, .tv and .me all placed in the top ten, supporting these as popular choices for businesses and domain investors looking outside the legacy big three.

Keep In Mind
All ranking systems are dynamic and change from day to day. I have been researching this article for a few weeks, and during that time the rankings were pretty stable. Also remember that any ranking system may mask huge differences between activity of different TLDs even when their rankings are only slightly different. On the other hand, different rankings may actually correspond to rather similar use rates, especially as one goes further down the list.

It is important to stress that these are rankings based on perceived global traffic. No doubt each country code extension finds extensive use within its own region, even though when measured globally many are well down the ranking list.

Have Your Say
I would love to hear your reflections.
  • What things most surprised you in the rankings?
  • How important is web traffic when you consider investing in an extension?
  • Which extensions do you predict are on the way up, and which are on the way down, in terms of web use?
  • In a year’s time what extensions that did not make this list of top 50, would you expect to now be on the list?
 
44
•••
The views expressed on this page by users and staff are their own, not those of NamePros.
@Bob Hawkes thank you for this interesting article, I have really learned from reading your articles, post and questions!
 
1
•••
w/o data.(ok, as a fakepro, I guess a lots things w.o data...u can just ignore)...but I just guess 90% web traffic is come from top 10000 big websites....if this is true, using the traffic to consider domain investment is not so meaningful.....may be the reason .news got such highre ranking just because 1 or 2 TOP site contribute most of the traffic while they use .news extension? I may need check what is the biggest .news site...and how big it is...

the real surpise for me is .cloud, I establish a .cloud disucssion thread, but no single response, thus I assuming no one really care this extension, but it even get a really good ranking here....I have no idea....if u have any idea, go to this thread to comments...Showcase & Discussion for .cloud

my expectation for future is, .one will get more and more exposure...it is such beatiful and elegant, I expect it can win some position....again, I have no data here......{my presonal "love ranking" for extension is .com, .one, .monster.}

finally, thanks, Bob for ur nice sharing....great investigation..

another thing I want point out, I think .tv is by nature will win good position, as most tv site consume lots of traffic while people using their service to watch video may be...I am not surprise on this....
 
Last edited:
2
•••
Thank you @Bob Hawkes for this article, very delicious to read while taking breakfast in Central Europe.

The ranking is really interesting. My thoughts:
  • There is finally a "real" ranking of a few nTLDs. Do you have numbers to continue the list, eventually?
    • For me as a domainer dot .club ranks as I would have expected.
    • The other handful of nTLDs mean nothing to me as an Internet user.
    • I have even never seen a live website in dot .news. Thinking about "news" being one of the most prominent words in multi-word domains it is understandable that there is some movement of content to the nTLD.
  • Better than expected are .io, .tv, .ms, .me, .ly.
    • Dot.ms - I cannot nearly believe that Microsoft and Mississippi are creating that much traffic.
    • Dot.ly I would never touch because of the chaotic political situation in Lybia. And I thought to watch that the extension is a bit out of fashion.
  • Disappointing are .de and .cn regarding the number of domain registrations.
    • Nearly unbelievable "weak" is .cn if you think of the pure number of people in China, despite some limitations by federal control. Perhaps there are more answers by @Kassey Lee ?
  • Worse than expected are .eu, .in
  • Somewhat surprised that com.au, co.nz and .cc are not within the Top 50.
 
1
•••
Thanks @Bob for letting us know the updated extension
 
1
•••
Somewhat surprised that com.au, co.nz and .cc are not within the Top 50.
I also surprise for .cc out of the list, it is another lovely tld in my mind...
 
1
•••
Surprised that .uk was above .co.uk
 
4
•••
Until this post, never gave it much thought, as most of our domains have always been .com, and developed as ad-free no-purchase sites for health and wealth self-empowerment.

Upon learning semi-accidentally ("No such thing as coincidence",: right? :) ) that unlimited # of excellent generic single words were available in .work, went a bit overboard. Godaddy was charging $1.99 each, and we seem to've burned that bridge by buying 900+ in one session, then another 7,100 in 2018 ... which they've now upped to $8 (oy). Understanding I'll have to rapidly build a thousand discretely hosted sites in next 50 days to see .work on your list by Jan, ey? (It's a statement as well as a domain!) Purple color to indicate responses to your 2nd, 3rd, 4th q's. :)

If I could understand WHY .io and .co are so high on the list, happy to buy some. So far, it's baffling

One big gift you've given in the lesson is .tv. We only have a few, but it seems to me that, if it ranks that high in usage, it's a HUGE advantage for those with a larger budget, considering how expensive they are. Off to go find some juicy one-worders in .tv extension. Thx

2) Having now read over a thousand-plus NP posts, it's clear who the Professor among us dummy domainists is proving to be. Several of your posts are instructive, utile. Repeated thanks for your generosity of spirit, and elevated teaching skills, replete w/ socratic technique rather than just shoving your opinions.

Would guess you've been teaching people all your life, meaning you must have some exeedingly successful students. Aiming to join that group. Best of energy to you.
 
3
•••
there must be something wrong

where is .de ?

the germans got way more than most of them
 
0
•••
  1. .com 114.669.959 registrierte Domains
  2. .tk 26.546.946 registrierte Domains
  3. .de 15.775.003 registrierte Domains
  4. .net 15.086.604 registrierte Domains
  5. .cn 10.906.655 registrierte Domains
  6. .uk 10.513.608 registrierte Domains
  7. .org 10.423.843 registrierte Domains
  8. .info 5.593.495 registrierte Domains
  9. .nl 5.506.314 registrierte Domains
  10. .ru 4.894.636 registrierte Domains
https://www.keyweb.de/de/keyweb/blog/2015/die-teuersten-und-am-meisten-verwendeten-domains

.de is No 3
 
2
•••
Quotes from your article:

“There are various measures of internet website use, with the main ones being Alexa, Cisco Umbrella, Majestic, SimilarWeb, and QuantCast.”

“The Cisco rankings are based on the number of unique DNS (Domain Name System) queries are made for a domain, but only counting those from different IP Internet Protocol ) addresses.”


End Quote.


So I have two questions, first what is the effect of static or dynamic IP addresses in Cisco’s ranking system and how do they account for that, and my second question is what results would you get for the top 50 TLDs if you took the average for each TLD of the combined ranking data from all the various measurement systems that you had indicated.

And thanks for a great article.
 
Last edited:
0
•••
The rating itself shows that it is absolutely unreliable. Whatever they are counting with IPs, it has little correlation to the traffic in the sense of visits by actual people to actual websites. Few domains doing some infrastructure services could tilt it hugely, as is probably the case with .ms and ly. And that is probably what is heavily influencing .io, .co...

.news above .de is a joke.
 
1
•••
Surprised that .uk was above .co.uk

My thoughts exactly - I just don't see that as being feasible at this time. Maybe in 5 years time but for todays stats never.

My guess is they have dumped everything ending in .UK together, then have taken an account for the .co.uk separately. Wonder what else they got wrong.
 
Last edited:
1
•••
My thoughts exactly - I just don't see that as being feasible at this time. Maybe in 5 years time but for todays stats never.

My guess is they have dumped everything ending in .UK together, then have taken an account for the .co.uk separately. Wonder what else they got wrong.
Surveyed the .uk subdomain a few months ago and the usage was not high. Does look like an error on Cisco's part.

Regards...jmcc
 
2
•••
I have been mainly offline today so sorry a bit late responding to comments (someone told me taking a break was a good idea :xf.wink:).

@Zilla your point that if traffic is dominated by big sites, which are on domains outside the range of the average domainer, is a good point. Although, I think people seeing real use being made of the extension would have some carry over effect in saleability of $$$ to $$$$ domains in the extension.

may be the reason .news got such highre ranking just because 1 or 2 TOP site contribute most of the traffic
I actually had wondered whether the 6 "news" sites that sold last year for $16,000 + each, and are operated by the same controversial USA person, might be accounting for it but it seems not so. These are not in any way based on Alexa but if you look at Alexa 1M stats for .news they do not rank nearly as high, but they are top 10 with 419 in Alexa 1M, or 1 for every 140 names registered, one of the higher ratios and actually higher than .com as a ratio. I also did a Google .site search and there seem a fair number of real news sites from around the world operating on the extension. So no, I don't think it is just one or two with huge traffic.

Re your .cloud comment, remember these are web use stats which is something completely different than value in domain aftermarket and domainer interest. I think use is not irrelevant, but it is only one factor.

Do you have numbers to continue the list, eventually?
Yes, I will do so in a separate post.

I have even never seen a live website in dot .news.
I admit I did not see many either, but I think it is the case that the ones that are out there are all getting a reasonable amount of traffic. When I first began researching the Cisco numbers a few weeks ago I was so surprised about .news and especially .ms that I held off until I looked at a few different listings to make sure that they persisted.

I wonder also if by the nature of .news sites, lots of references from social media, get lots of people looking repeatedly and casually at them so the numbers add up. Although I would have thought .info and .edu would have been higher for the same reason.

I also speculate, could be totally wrong, that perhaps the .cloud and .io strongish numbers are due to the nature of the types of resources at those sites - like lots of people getting bits of code from an .io site or cloud based data from a .cloud. I have no idea whether this reasoning makes sense.

Disappointing are .de and .cn regarding the number of domain registrations.
At first I was surprised but I think we need to take into consideration that .de has a population of about 83 million in a world population of about 7.7 billion, or about 1%. No matter how active it is in its region, it is hard for regional domains to complete in global traffic with ones that have some traffic from all over the world.

I suspect that there is some bias against Chinese traffic, maybe a lot, in the data, and oppositely probably a pro-US bias (even though Cisco is very global) and that might explain why .us did better than I would have thought, and .cn not as well.

Somewhat surprised that com.au, co.nz and .cc are not within the Top 50.
I was surprised that .ca that is so dominant here was not higher, but then when I thought about that it is almost entirely Canadian and our population it makes sense that in global traffic does not do so well, and I think same for au and nz (I think they were not far off - will give more data shortly). Re .cc, not sure. It certainly seems to sell better than many others.

Surprised that .uk was above .co.uk
Well they were side by side which could mean virtual tie, and probably does. I don't know the details of the co.uk to matching .uk transition, and others who know .uk have already provided comments.

@Michael Ehrhardt the stats you are giving are registration numbers. This is data on traffic, at least as estimated by domain name calls from different IP addresses, and does not directly relate to registration numbers. There is no doubt that .de is a very valuable and highly used in Germany extension. But as I mentioned above, with 1% of world population not surprising it will not be at top in global use stats.

first what is the effect of static or dynamic IP addresses in Cisco’s ranking system
It is a good question and I don't know for sure, but I would think as long as the calls came from different IP they would be counted, so it would open up ways to try to manipulate via dynamic changes, probably. But I don't know, just speculating.

Understanding I'll have to rapidly build a thousand discretely hosted sites in next 50 days to see .work on your list by Jan, ey?
HaHa :xf.grin: Well you could try to modify the stats that way, and I am sure someone does, but with so many global calls better get a big number of those discretely hosted sites!

If I could understand WHY .io and .co are so high on the list, happy to buy some. So far, it's baffling
Really .io does have a big following in the tech/code world. I was a little surprised to see it after the legacy three, but remember that the ranking may be a tiny difference, and I suspect that it is. I suspect .com is like 8x more than .org and .net and they are about tied, and then another fair jump before the next few that are close together. I am not saying this directly from data, but I suspect that is the case. I was not surprised to see .co high, in fact expected it a bit higher than .io. For both the issue as a domainer is really the renewal costs that are like $25 ish.

I will get through responding to the other comments, but think I will post this one now.

Thanks for the great questions and critical thinking everyone.

Bob
 
3
•••
Before I get back to the last of the comments just wanted to stress a few things....

Remember that the differences may be very slight, like we don't know but places 6 through 50 may all be almost the same. So I think we should not get overly focussed on reading things into possibly small differences. Yes, it is interesting which is where, but we should not read more than a ranking system is telling us for sure.

Keep in mind that this is one way to measure web traffic. As we all know Alexa has been criticized because it can at essentially zero cost be manipulated, but every other system can be manipulated too. I think the fact they insist on different IP addresses makes it a bit (but only a bit) more challenging to manipulate but each has its weaknesses. In the research article linked in my OP they discuss how to manipulate each of the systems. Every system can be potentially manipulated. Overall, I like the huge dataset that the Cisco is based on, and the fact that CENTR (who I respect a lot) have begun incorporating it speaks to its quality, but it is not perfect for sure.

While use is I think relevant to quality of an extension, the value of a domain involves many factors so a domain that is more highly ranked on Cisco is certainly not necessarily a higher value domain, in general.

Despite the criticisms of it, Alexa is probably most widely used measure in our industry. Keep in mind that how the rating is done is very different between Alexa and Cisco in two ways...
  • Alexa counts (supposedly human) visits via an installed browser plug-in. The majority of users never install it, and it is not even supported for many browsers. There is a danger with Cisco of manipulation without humans, but at least it is counting calls to domains from different IP addresses.
  • To look at an extension data in Alexa one usually does it by how many are in the top 1M. But that is not the same as how many calls to all the TLDs in an extension are made from separate IPs.
Bottom line is Alexa and Cisco do it very differently, and not surprisingly have somewhat different results.

Bob
 
1
•••
my second question is what results would you get for the top 50 TLDs if you took the average for each TLD of the combined ranking data from all the various measurement systems that you had indicated.
The perfect question. OK the simple answer is I don't know because it is not readily done. Let me explain...

The reason I used Cisco Umbrella, and I suspect the reason CENTR have started using it, is that it provides (free) two lists, one of domains ordered in use and one of TLDs in order of use.

The other methods, to my knowledge, do not provide the TLD list so people derive one by counting how many of each TLD are in the 1M list. That is a logical approach, but is not the same thing. If one extension had a couple dominant sites with all the traffic, another had many moderately high traffic, and another had lots of traffic spread over essentially all of the domains, not just the 1M, they would get very different rankings in Alexa but might be very consistent in Cisco Umbrella.

Anyway there is not an easy way, for TLDs, to do what you ask.

TRANCO which I mention briefly, tries to do exactly what you ask for domains however. They have an algorithm for combining data from Cisco Umbrella, Majestic, Alexa and QuantCast. They are trying to make manipulation difficult, so since it is different ways to manipulate each, if they take all four into account and a 30 day period which they use, one would really have to invest a lot of sustained effort to manipulate. I think more and more will use TRANCO in future, but as of now it is a domain, not a TLD, list.

Bob
 
1
•••
Few domains doing some infrastructure services could tilt it hugely,
I agree that is certainly possible. There is a good writeup in this article I cite above about how each of the four systems (they do not cover SimilarWeb) could be gamed.

.news above .de is a joke.
It is true that in Alexa 1M .de is above .news in absolute numbers (although no where near it in proportional numbers). I think we need to keep in mind that with main use in Germany, with 1% of the world population, even with many global German companies (although most are on .com I think) it is hard for a regional domain to get in the top 10 of global traffic.

It does seem to me that .news has a fair amount of significant traffic spread over many domains. I know many do not like Alexa based use ratings either, but I had a look after this discussion on what W3Tech (that use Alexa as starting point) have to say about .news. They do plots showing how heavy use is vs how many websites on it. Here is the graph for news , so very small number of sites, but the ones there are high traffic. .net, .org, .com, .de, .ru are shown for comparison. So this is totally unrelated to Cisco Umbrella but suggesting high traffic per site.
NewsW3Tech.png

Thank you for your comments.

Bob
 
1
•••
Great article. .com .org .net remain on top, do you think another extension will ever take the 2 or 3 spot?
 
1
•••
The thing about .News is that with so much News happening all around the World people repeatedly go to News websites all day long and so it can add up to a lot of traffic specially if they use dynamic IP and each visit counts as a new unique visitor. IMO

Also as far as .US is concerned a lot of schools and Institutions use this extension which amounts to a lot of traffic. IMO
 
Last edited:
2
•••
My guess is they have dumped everything ending in .UK together, then have taken an account for the .co.uk separately.
Thanks for the suggestion which I agree is a possible explanation. I really don't know enough about the UK market to comment more than that. I tried to see how the relative Alexa numbers were for the two, but unless I am missing it https://domainnamestats.com/ just seem to give .uk stats and not co.uk separately that I could find?

Does look like an error on Cisco's part.
Thank you.

Repeated thanks for your generosity of spirit, and elevated teaching skills, replete w/ socratic technique rather than just shoving your opinions.

Would guess you've been teaching people all your life,
Thank you for your very generous comments, @bidigitals :xf.smile: Yes, I have been, one way or another, teaching all my life, from toddlers to senior citizens. I feel very fortunate to have had that opportunity. Re rather than just sharing opinions, I am glad you feel that. The goal I always have. is to in as balanced a way as possible, and as supported by evidence as I can find, present the case, but everyone really does need to make all of their own decisions.

Like just to be clear I have never owned a single .news domain name or even looked for one, nor an .ms or an .io. I was surprised when a few weeks ago some of the data from the Cisco Umbrella TLD list, so thought it would be worth sharing. Thanks again for your overly kind remarks!

I really appreciate all of the care people have taken in expressing their views.

Bob
 
Last edited:
1
•••
Overall, I like the huge dataset that the Cisco is based on, and the fact that CENTR (who I respect a lot) have begun incorporating it speaks to its quality, but it is not perfect for sure.
Centr also uses the numerology from NTLDstats on "parking", Bob,
This is the "parking" stuff that was showing LOAN as having 30% non-parked domains until the gTLD imploded recently. W3Techs is also highly limited to the top 1M Alexa sites.

The rankings above just don't look right. The MS ccTLD ranking may be due to Microsoft telemetry and updates. CO and IO just are not in the same league as the legacy gTLDs either in terms of active sites or usage.

The data, from what I can see, is based on data from its products rather than data from the root servers or registry data. That means that it will not necessarily be representative of general trends especially where Cisco Umbrella does not have a major share of the market for such products/services. This is probably why it massively underestimates ccTLDs. The .uk versus .co.uk rating is just plain wrong. There isn't that level of overlap with the .uk websites and usage and redirects are still very low. Many of them are pointing to a registrar holding page.

In terms of NET/ORG, ORG should be ahead due to the use of Wikipedia.org. A generation of schoolkids has grown up using it for homework and it is a very heavily used site. That's why Google scrapes it for its "knowledge" box.

Regards...jmcc
 
0
•••
I agree that is certainly possible. There is a good writeup in this article I cite above about how each of the four systems (they do not cover SimilarWeb) could be gamed.

It is true that in Alexa 1M .de is above .news in absolute numbers (although no where near it in proportional numbers). I think we need to keep in mind that with main use in Germany, with 1% of the world population, even with many global German companies (although most are on .com I think) it is hard for a regional domain to get in the top 10 of global traffic.

It does seem to me that .news has a fair amount of significant traffic spread over many domains. I know many do not like Alexa based use ratings either, but I had a look after this discussion on what W3Tech (that use Alexa as starting point) have to say about .news. They do plots showing how heavy use is vs how many websites on it. Here is the graph for news , so very small number of sites, but the ones there are high traffic. .net, .org, .com, .de, .ru are shown for comparison. So this is totally unrelated to Cisco Umbrella but suggesting high traffic per site.
Show attachment 131534
Thank you for your comments.

Bob

That is also a little useless chart and here is why:

judging by the position of .com, it is averaging all .coms in the rating, and of course, it pulls down the average a lot.

Given that there lots of .de names there, there average is also bound to suffer and show even lower than .com

For news, probably few made it into the 1 million and their average (or mean) shows much higher. It is ridiculous to assume that .news sites even for media are above .com. All top media outlets use either .com or .org or their respective cctld.
 
1
•••
Thanks for the suggestion which I agree is a possible explanation. I really don't know enough about the UK market to comment more than that. I tried to see how the relative Alexa numbers were for the two, but unless I am missing it https://domainnamestats.com/ just seem to give .uk stats and not co.uk separately that I could find?

Thank you.
From the UK zone from 01/October/2019, the .uk sub was at 2,163,897 domain names. (05:14 here so the grep/regexp might not be exact.) The co.uk sub is still the largest in the ccTLD and most of the usage is in the co.uk as it has been around decades longer.

Regards...jmcc
 
1
•••
CO and IO just are not in the same league as the legacy gTLDs either in terms of active sites or usage.
And this ranking is not saying that they are. It is a ranking system, the difference between places 1 and 2 may be huge (which I am pretty sure it is), and the difference between places 3 and 4 may be large (which, not sure, but suspect). I said this a few times, but it is important to not interpret more than the data is telling us.
The data, from what I can see, is based on data from its products rather than data from the root servers or registry data. That means that it will not necessarily be representative of general trends especially where Cisco Umbrella does not have a major share of the market
Yes, I agree and mentioned that almost certainly .cn and some other countries are under-represented due to this. Nevertheless they do list all the locations where Cisco servers are located, and it is pretty global, and 100 billion served requests per day is significant. But I agree that it, like all data, has biases.
Also as far as .US is concerned a lot of schools and Institutions use this extension which amounts to a lot of traffic. IMO
I had not realized this. That probably does explain a lot of it.
The thing about .News is that with so much News happening all around the World people repeatedly go to News websites all day long and so it can add up to a lot of traffic specially if they use dynamic IP
This seems exactly right to me. I think the Cisco way of measuring estimated traffic will favour things like news services or information services.
Great article. .com .org .net remain on top, do you think another extension will ever take the 2 or 3 spot?
I think it is always possible, but don't expect it to happen anytime soon. If I had to bet on one to do it, probably would be .co. They seem to have been really active in promoting the extension, has a pretty solid history, and with the big promotions in July the base must have gone way up. I seem to see more of them than I used to even a year ago in real use.
In terms of NET/ORG, ORG should be ahead due to the use of Wikipedia.org
I agree and actually expected .org to come out ahead even though .net has more registrations. I suspect they are almost tied. In terms of how many Alexa 1M .net is 34,048 and .org 33,250. I agree with Wikipedia a top 10 overall site it takes a lot of moderately popular ones to make up for it. I am pretty sure that within the year .org will surpass .net.

An important factor that I realize I don't know is how redirects are handled. I presume someone going to a redirected domain still counts as a count for what you initially went to, but I am not sure. It might make a big difference on some. For example Apple.news redirects to their news service, but not sure if people are using that or if Apple have promoted the .news in a campaign. But overnight they could change traffic a lot. Amazon.news also redirects to a news section. Even here in Canada I discovered that CBC.news redirects to our national "public" news service, a huge network.

Thanks for all of the comments, everyone!

Bob
 
Last edited:
3
•••
Back