Shared publicly  - 
 
I read a post by someone offering new top-level domain (TLDs). They made this claim: "Will a new TLD web address automatically be favoured by Google over a .com equivalent? Quite simply, yes it will."

Sorry, but that's just not true, and as an engineer in the search quality team at Google, I feel the need to debunk this misconception. Google has a lot of experience in returning relevant web pages, regardless of the top-level domain (TLD). Google will attempt to rank new TLDs appropriately, but I don't expect a new TLD to get any kind of initial preference over .com, and I wouldn't bet on that happening in the long-term either. If you want to register an entirely new TLD for other reasons, that's your choice, but you shouldn't register a TLD in the mistaken belief that you'll get some sort of boost in search engine rankings.
183
114
Jake Weisz's profile photoDietmar Stefitz's profile photoGavin Farrington's profile photoKevin Mullett's profile photo
75 comments
 
Thank you for putting that to rest. :)
 
Will keywords in new TLDs have ranking impact? For example, if Coca-Cola registers ".coke" and then puts up a page at "buy.coke", will the ".coke" be treated as, say, a sub-domain ("coke.") might, keyword-wise? Traditionally, you don't think of TLDs as carrying keyword weight, since you don't generally search for "com", "co uk", etc., but I just realized that I have no evidence either way.
 
This sounded like a lie the moment I read the claim. I think it's almost common sense that it's false. Unfortunately, common sense is so rare, it's a super power.
 
Thanks +Matt Cutts. I wouldn't trust anyone who spells 'favored' incorrectly, either. ;)
 
+Pete Meyers I kind of doubt Google actually cares at ALL what the domain is, if it's a page Google can find, index, and link to, and someone finds it via searching for relevant text, Google's going to throw it out there regardless of what it's URL is.
 
Unless you use a site: filter, of course.
 
Yes, content, content and something compl. different - content!
 
"Adrian Kinderis [author] is an internet industry thought leader and CEO of ARI Registry Services, a top-level domain specialist with the experience and technology to activate and implement new top-level domains."
 
I do enjoy how SEO-marts try to sell themselves as somehow knowing the 'secret-sauce' on how to get higher ranking, w/o real content.
 
I've got to assume the TLD will be added in as a signal. Given the expense of registering a new TLD, it's possible sites with new, custom TLDs will be of higher quality than most .com's.

However, even if that turns out to be true, it's no guarantee that Google will favor custom TLDs. The effect on relevance might go away after controlling for other factors.

The best strategy to rank #1 for any term will always be to create the best site on the web for that term.
 
Thank you for keeping us in the know, Matt!
 
I think it'd make sense for Google to favor certain TLDs in certain cases. Such as if "pepsi" bought .pepsi; results for pepsi related queries would be higher. That'd make sense, imho.
 
+Navarr Barnier so, if I bought .cars, I should have preferences on searches for cars? With widespread TLD creation, it doesn't make sense to judge which ones should or shouldn't get a bonus. Google ranks sites on content. I would imagine Pepsi's site has the best content about Pepsi, they don't need a special bump.
 
Jake Weisz: Thats not what I was implying. If PEPSI bought Pepsi it should have a higher preference, but not some generic TLD bought by a third party that doesn't matter.

.aero websites, for example, should be featured over .com - as .aero is regulated aeronotics. That's the point I'm making. That regulated domains that have high limits on content should be favored over TLDs that aren't. But not generic TLDs that aren't at all regulated.

so, if you bought .cars but it was under contract in such a way that only automobile companies could purchase their trademarked names, then yes it should get a bonus - is my point :p
 
To be fair, what the article actually says: On a given page of results, users might be more inclined to trust and click a .brand TLD. So even if Google doesn't rank it higher, when a user is picking from among the (say) top half dozen results, they would be more likely to pick the .brand one.

I don't know whether I agree with that, but I figured I'd point out that this is actually his main claim (not that the TLDs will directly affect Google's ranking per se) so you all can explain why that is wrong. :)
 
To correct myself. He's really making two claims. The weaker claim is that Google will rank it higher. The second claim is that users will click it more often. Even if Matt says the first is incorrect, there's still the second one which I think is interesting even if I don't know whether it's correct.
 
+Andreas Roedl They already can, but they have to manage and host it, and it's not cheap.

+Navarr Barnier There's no point. If the content is good, Google will rank it well. If people frequent the content, Google will rank it well. You don't want to give preferred rankings for anything besides content and popularity. If you do, you're making it possible for people to buy better rankings, more or less. Even if it's limited to certain people, actually worse then because it's favoritism. Google's entire war against SEO is about ensuring that no special treatment is given to folks based on how much money they fork out.
 
+Jake Weisz makes some good points, but I disagree with a premise - google doesn't rank good content, it ranks popular content. The quality of the material is (still) practically irrelevant without the popular factor (a tree falls in the woods with no one to hear it = a page gets published online but no one links to it). 
 
I think it's funny that the article very clearly says that these new TLDs will be better for SEO, but then it buries the disclaimer in the fact that the TLD would need to build a sense of community and trust surrounding it.

In other words, it has nothing to do with the TLD itself and everything to do with any other moniker to indicate trust to the user. For instance, if you trust Wikipedia for information, you will seek out that link instead of others. If you trust Experts-Exchange, you will seek out those links. So, the only way the change of TLD is going to have an impact is if the TLD first builds a sense of trust with users. That's just a no-brainer and has nothing to do with some sort of secret sauce or Google algorithm, yet the article is framed to make it appear so.

I also think it's going to be an uphill battle to get the common user to trust other TLDs beyond the currently popular ones. It will take a lot of marketing power and there's still no guarantee of success. Anything that looks different from the norm will automatically look like a spammer's trick.
 
+Matt Cutts I appreciate the reply (and the video), and perhaps I used "popular" incorrectly. My perception (humor me:) is that you use a multitude of factors to evaluate the quality, relevancy and authority of pages/urls, the majority of these factors being "off-page". I don't think Google "just" ranks for any reason, but I would be surprised (and pleased) to hear from you that content quality (which I assume is primarily on-page factors) is a stronger factor than popularity (which I'm hastily re-defining as "link popularity", a major off-page factor), when it comes to SERP positioning/ranking (and not just PageRank).

If content is published and not popular (few shares/links to it), does the quality really matter, or is quality only a factor among the most popular content?
 
+Aaron Douglas it's often a trade-off between reputation and topicality, but we do care a lot about topicality, not just reputation.
 
Thank you Matt, such comments are highly expected from the new gTLD industry. May I ask the question in more simple words as I doubt people read long articles: If I should launch a web site in 2013 (or later when the .wine registry is launched) which content is about wine: would google favor the one with a .wine extension or the one with a .com extension?
 
new TLDs could be nice to have. But it's good to know this... coming from +Matt Cutts
 
Now if Google would kill this ridiculous exploited back link rating and use true content, we'd have us a search.
 
It was obvious that it was an article with an agenda. Thanks for posting about it right away Matt. The new Tld consultants should focus more on how these extensions will gain mindshare imo rather than talking about search ranking in Google.
 
Do we still believe that say a 2 yr registration of a domain shows quality over the 1 yr registration since spammers have the habit of hopping domains quickly as they are banned?

One thing I can tell you for sure is that brand new domains are now pulling high Spam Assassin scores in email newsletters. So that brand new domain of your is not going to allow email to even a double opt in list like mine to even get to a spam folder.
 
+Matt Cutts While I appreciate the clarification and good news that .TLD's will not be weighted, ala .cars, .realestate or heaven forbid .seo, what I don't hear is rational for the weight of EMD (exact match domains). i.e. or the favoritism towards brands. (Yeah I know...you've heard this from me before.)

To be fair it appears as though this has been scaled back recently (again), but to simply repeat the mantra handed down that content rules, as some here are doing, belies evidence that small local businesses are frequently beat out when someone buys a cityname-keyword-service.tld.

Small local businesses should not HAVE to hire an SEO to win over a word press blog with content that says "Welcome to WordPress. This is your first post. Edit or delete it, then start blogging!", that has an EMD. Small local businesses should be able to rank for the service they provide and describe on their site by using the brand name that is on their sign as their domain name. I maintain that everything to the left of the .TLD/.gTLD should be neutral. Use the paths as a signal if you wish, but leave the .TLD's out of it. Domaining and squatting would become less of an issue and then you would truly be ranking based on content. Please and thank you.
 
Just what I have been saying in open discussions with iCann and some of the domain administrators. Just look at a search for "museum". There is no sign at all that a .museum site would have any boost what so ever over another TLD. (And the guy running the .museum TLD went for my throat in a way that clearly stated that he new nothing about search whatsoever.)
 
Even if all that were true, the benefits of that would still be apparent through the measures that are already used. Hitting a few invented bureaucratic check marks does not automatically mean it is any more relevant to the user, and if it is more relevant to the user then you'll see that through inbound links, shares, and other content relevance measures.
 
Funny, I get messages and phone calls about this all the time. Now I can confirm this and point them to this post. Thanks Matt.
 
The best experience for the user is what G focuses on. If you've got a nice page that addresses the user's needs, G wants to show it, as it should.
 
Also - the potential benefits that most people are discussing here are no different than an existing trusted TLD or CCTLD - just semantics. Which brings us back to square one - what new value and novelty does a new gTLD bring?
 
Hi +Matt Cutts . You are dispelling any misconception that a new gTLD is going to automatically jump you to the top of searches, but is Google actually going to ignore the new gTLDs?

I'm sure that the domain name does have a strong influence on the search results - at whatever percentage of the algorithm. If you completely ignore the gTLD, then surely every new gTLD is at a large Disadvantage?

As an example, somebody may own http://www.sheffield-wine.com, and Google would acknowledge the two words "Sheffield" and "Wine" which is quite specific. If, however, they had http://sheffield.wine and the gTLD was completely ignored then Google would only consider "Sheffield", which is nowhere near specific enough.

Similarly, www.tesco.com/shopping www.tesco.com/insurance, etc would have a major advantage over something like http://home.tesco or http://insurance.tesco. These kind of domains are the key for branded gTLDs (as well as specific addresses for far more measurable marketing campaigns). Surely Google will not ignore them?

I think that the new gTLDs is great news for search engines. At the moment, the internet-savvy will often guess at an address, e.g. [www].brand.[com/ccTLD]. With the multitude of gTLDs it will not be so obvious what address would be and the younger generations will be "trained" to always search for the website they are looking for.

Assuming that you are going to consider the words within the gTLD, then these will be an improvement for companies who want a relevant domain name and are starting out or that had to settle for something obscure or a completely random name, which is all that you can find on .com these days.
 
+Gary Barclay basically, keywords to the right of the dot will still have to earn their search visibility the same way as those domains with keywords to the left of the dot - which is with good content, links, etc.
 
+Matt Cutts thanks, but...there are so many people who lie on many other things about Google, and Google does not say anything... why don't you do a "True or false quiz" webpage to help people understand?
 
+John Mueller thanks, I did it (in Italian) .I think it could be useful to create a quiz "True or Fals" to ask people what they REALLY know about Google, you'll find a lot of interesting things. For example, in Italy there's someone who thinks that if he clicks on ads then he'll pay!!
 
+John Mueller I often ask in Google webmaster forum, but I never find an answer to my question. Often the people answering are not so skilled about seo topics.
 
Many SEOs still confuse cause and effect, correlation and causation. The most intelligent conclusion that they can make about .edu domains being visible for academic topics, is that somewhere in Google's code there must be a condition like "if the query is about an academic topic, then give more weight to .edu domains.". Obviously, these people know nothing about computer science and information retrieval.
 
It's unfortunate that a misconception like this got the publicity it did in the first place. I may still have a bit to learn about what Google appreciates (and what it disregards), but this TLD-for-instant-optimization idea smacked of goofiness right away. Nonetheless, thank you for your input, Matt. I like hearing remarks like yours directly from the source.
 
+Rob Garner - Thanks. I read through the original article and it seemed that, apart from his conclusion of "car.insurance being better than carinsurance.com" being wrong, most of the points are still valid. By that I mean that if you were to attain a more relevant domain on a new gTLD rather than a less specific .com address, it would be better. As an example, if you were a lawyer, www.robgarner.law should be slightly better than www.robgarner.com because of the keyword in the domain.
 
Thanks for the clarity Matt. I think that, in general, if a TLD is used to better brand a business it could still have an indirect impact on search...just a thought!
 
A problem in the SEO industry - Google seems to be ANTI-SEO, so maybe we can rebrand SEO as something about making great sites that get 'great rankings' (content strategy experts [CSEs], or Makers Of Great Sites [mogs?]...hmm how about "Make All Top Tier[matt], nah failing hard there...... So the problem i see with SEO info online is that the people who know the answers are smart enough NOT to post them online that often. However, sometimes this is a good thing because there are many out there who claim to be SEO experts, but simply google the things they are trying to do when clients ask them (we're talking the under $500 / month range of clients that expect page one results and the people who take them on) </end rant>.

hmmm (@KRONiS)
shameless music plug: http://soundcloud.com/kronis feel free to burn me if you think it sucks.
 
That would be too easy....and those who have big budgets would outrank good quality sites. Thanks Matt for straightening out this myth publicly!
 
O Matt deu um banho na estima do Adrian Kinderis. Excelente aclaração, Matt.
Translate
 
I agree with Aaron. Clearly the second anything SEO related is released or talked about by Google is the second it isn't as impactful. Don't get me wrong, it is still great to have the resource to have!
 
I appreciate ICANN made its 43rd meeting in Costa Rica, it was a wonderful experience for all of us that participated, the new TLDs in many parts of the world are considered a very strategic opportunity to integrate services and regions by the governments in charge, and therefor they are going to be part of this new universe and new order in the online community, for some parts of the world still in development and that is very important for us as internet pioneers, we should understand the importance of having a way to preserve our style of life by all this new labels seemingly small for some but very relevant for others, the world is getting smaller and creating "mind parks" we all understand and are familiar to us in our languages and cultures should go forward the same direction, I know as a fact that I will love to be able to understand the internet because I am able to read it, I know as a fact this change might create domains I would not be able to read and it will reinforce local preference for tools and languages in many regions, something is very delicate and very important to analyze before tendencies start to arise, the new TLDs can mean a stop or an opportunity depending of how you look at it, as I said we should focus on how to be able to continue our growth uniting the world with tools and the Internet, which it has let us shown the world the advantages of living in freedom and modernity, which is something many countries does not have, very important to think about that, of course this is just my opinion, and again thank you to all the community for visiting Costa Rica and showing all this interesting facts about the Internet for all of us learning our way up ;)
 
Thanks for this information matt.
 
Thanks 4 ur information, Matt :)
 
Why aren't ccTLDs allowed to use "Unlisted" option in Google Webmaster tools (Geographic target)?

I think that creates a disadvantage for these sites against sites that don't use ccTLDs.
 
well if you check the TOP 10 listings for education one would expect to find .edu domains. In fact non .edu is listed which is pretty much the same with the new endings.
 
But the string "edu" is not necessary read and understood as the word which defines education, same for "aero", same for "info". The string "museum" is the correct word to qualify a museum as such.
 
Thanks Matt, this clarification will put alot of people at ease
 
Thank you very much for your enlightenment Matt Cutts
 
Not sure why this just popped back in my alerts, but just so everyone knows, new domains of any type get major scrutiny from URL filters, security software and spam filters.

I am currently creating landing pages within my site rather than the new domain of the week.
 
when new TLDs go on sales @ Google?
 
+Zev Mo Green favoured is the UK spelling of the word and is actually correct in the English language!  You are spelling it the US way - both are correct.  Just an FYI :)
 
Hi,
I’m putting this to the test.

I developed a directory site for a niche industry, including some snippets of microdata, fully mobile, with memberships on the site authenticating to google users.

The site has relatively OK SEO, social media integration, and focuses on search terms that it “should” rank for not because of the tld name, but because of the content. http://trusted.photography/ Let's see...
 
It's good to hear Google talk about the longevity of the domain name. 
Add a comment...