Only if it improves search results.
Using a new top level domain name instead of a .com or other gTLD does not give you a benefit in Google search results. Ditto for using a .com instead of a new TLD. They’re basically the same.
That’s what Google says. And it makes sense when you think about it. A lot of people that think new TLDs have an SEO advantage are thinking with their heart, not with their head.
Logic dictates that Google would give a boost of some sorts to domains on new TLDs if it made its search results better. I can’t think of a plausible argument at this point in time that Google SERPs would be better if new TLDs received some sort of boost. If Google’s algorithms accidentally gave a boost based on TLD, it certainly would scramble to fix this.
While it doesn’t make sense for Google to favor a new TLD over an existing TLD at this point, it could make sense in the future. It would make sense if it improved search results. Let’s go through some of the arguments that ranking new TLDs ahead of other TLDs would benefit search results, and consider if they are valid now and if they will be in the future.
New TLDs Cost More
Low-priced TLDs tend to attract a lot of bad actors. If you’re going to use a bunch of throw-away domains for phishing, malware, or crappy content, you don’t want to spend a lot of money on them.
Conversely, do higher-priced domains eliminate bad actors? I suspect there are fewer “bad” sites on higher-priced TLDs. That said, there are already a lot of expensive TLDs out there, such as .co. I haven’t seen any suggestions that .co domains get a search engine boost because of their price.
A more likely scenario would be that sites on a low-priced TLD would be scrutinized by Google. Back in 2008, Google’s Matt Cutts predicted:
“A top-level domain (TLD registry) will offer domains for under $4. The result will be another TLD blighted by spammy domain registrations.”
Later that year, .info domains (often offered for 99 cents at the time) were suddenly (and temporarily) purged from Google. I don’t think there was ever an official response, but it certainly led many to believe that Google was tinkering with its algorithms because of problems with bad sites on .info domains.
Google is pretty good at snuffing out malware and phishing. A bigger issue might be sites with low quality content on low cost domain names.
Another argument from people in the “new tlds cost more” camp is that Google started using HTTPS as a ranking signal in search results. SSL costs something, the argument goes, so Google found this to be indicative of better quality sites.
It’s important to think about why Google did this. Here’s the company’s blog post about it.
According to the company, Google wants to change behavior to make the web more secure, and also make sure sites that Google links to are secure. But they also note that they tested the impact on SERPs to make sure it improved before making this change.
Is it that SSL certificates cost something that search results would improve (or at least be the same)? Or is it because sites using it are often higher quality?
Another plausible idea: in theory, sites that use SSL certificates have had their identity validated. Which brings us to the next idea…
Regulated namespaces have better content
If a TLD has regulations limiting who can use the domain, will it make the sites using the domains better?
Let’s use .archi as an example, because it comes to mind as a fairly restricted TLD. .Archi is short for .architect.
It doesn’t seem like it would make sense for Google to boost .archi sites over other TLDs in search results right now.
First, there are only about 1,700 .archi domains registered. A fraction of these are probably in use. There would have to be a lot of websites under a particular TLD for Google to bother putting it in their algorithm. There are a lot of great architects websites on .com, .de and other TLDs right now. They’re unlikely to switch.
Second, for there to be a benefit in search results, Google would have to be having a problem with “fake” architecture sites. I just searched for “Austin Architect”, and Google is doing a good job of presenting well-respected architects in town. I don’t see anything that looks troublesome in the search results. Validation isn’t necessary, it seems.
If Google were to implement some sort of bonus for .archi sites, it would also need to differentiate between types of architects that qualify for .archi domains and those that don’t. For example, if someone searches for a “systems architect”, they probably aren’t looking for someone who qualifies to use a .archi domain.
So in the future, restricted namespaces could rank higher. But a number of things would have to happen for it to make sense for Google to take it into consideration.
New TLDs signal what kind of content is on the TLD
The idea here is that a domain that ends in .photography tells Google that the site is about photography. A domain that ends in .tattoo tells Google that the site is about tattoos.
Does Google currently have a problem classifying content on a website? Do they need a domain name to tell them? I don’t think so.
Now, there is the whole “exact match domain” argument. Google has cracked down on low quality sites using exact match domains, but I still believe people when they say using a domain name that exactly matches the topic people are searching for gives you a benefit. If nothing else, having those words bolded in the URL in search results should boost click-throughs. But you can exact match to the left of the dot on a lot of different TLDs.
Geo domains should rank better for local searches
Google doesn’t yet use geographic new TLDs as a ranking signal. It does use country code domain names as an indicator of the intended audience of content, however. It also allows people to tell Google their intended geographic target using Webmaster Tools. These are just a couple of the factors considered in the location equation for search results.
It would seem to follow that Google would, at some point, consider that a domain is on a .NYC domain when responding to searches. I’m not sure if this would give sites a leg up in the SERPs, though. If you search for movies and you’re in New York City, Google already does a good job of showing you results that are relevant to your local area. The TLD can be just one part of this.
Google certainly doesn’t want to show a .nyc domain name to someone searching for movie times in Kansas City.
I think adding a Geo TLD into the equation on when to show a site or not makes sense, along with the other tools Google already uses to figure this out. I suspect Google will examine this in the future. Its criterion will be simple: if we take this into consideration, does it improve our search results?
In summary, I think the key to think about is if Google can measurably improve its search results by taking into account the top level domain name used. If it can, it probably will consider them. If it can’t, it won’t.
Website Value says
I personally think it can and it has an impact on SEO. For example my website valuation website (it is based on new gtld domain) is now ranked in top 30 results in all search engines for keywords like ‘website value’ and ‘website worth’
Matt says
Wow, Andrew, you give Google a lot of credit! With the kind of machine learning algorithms they’re using it is normal and indeed unavoidable that the algorithm is imperfect, and often has unintended consequences, especially with second- and third-order effects. Will they really “scramble to fix” something like new TLDs offering an unintended boost? They haven’t “fixed” PBNs, expired domains, paid links, barnacles, injected links, guest posts, or any number of other more serious issues, some of which have been around for over a decade and Google has explicitly declared war on them. Mathematically, the “No Free Lunch” theorem dictates that each of their optimizations in one area leads to deficiencies in another area. “Fixing” the influence of keywords in new TLDs would require tweaking the influence of anchor text and user behavior, two of the most sacred elements of the algorithm. It’s MUCH easier for them to say “TLD has no effect on search engine results” (which is true in the direct sense), and hope enough people believe it that there’s no widespread abuse.
Robbie says
The main concept people do not understand about GTLDS they can never go mainstream. Unless ICANN lets you register .ANYTHING you want, GTLDS will always be held back.
I had a domain sell KeywordSponge.com, now will there be a .sponge in the future most likely not, for this reason gtlds are not a clear cut solution, just an alternative for some. Really they will will always be reaching, but never quite get to the top.
Frank Schilling says
destinymoody.com is outranked by destinymoody.sexy .. search “destiny moody” .. same site owner
Andrew Allemann says
I show .com at #3 and .sexy at #5.
americandotorg says
Frank, remember that SERPs are personalized. Google is the preeminent Yes man, showing people what it anticipates they’re searching for.
Since you have an ongoing interest in Uniregistry TLDs, Google probably gives .SEXY websites (especially those you’ve visited or searched for already) a ranking boost for you personally but not for others.
This is the “filter bubble” effect.
Exactly the same misunderstanding arose in a Dec 6 post that Michael Berkens wrote about rankings for Coffee.club.
couponpages says
The wildcard here is that search engines don’t give every visitor the same results anyway. Two people, living next door to each other can search for something and the result page from both could be 100% different.
The hard truth is that while search engines originally were designed to find the best results based upon things like content, URLs and keywords, they now factor in so many other variables that you never know for sure where any given page will rank… at any given time. The #1 position for you at 1pm could drop to #50 by 5pm, because of an influx of social media buzz that you have no control of.
In the end, I think the true power of a good domain is simply how it sounds… to the public. If the public can trust and remember your domain, the search engines will too.
Right now, the public favors .com domains, so those with .coms rank higher than a similar quality site, with the same domain with .net or anything else. We can debate which comes first… whether the search engines ranked the .com version higher, therefore more people found it… or if the public trusted / remembered the .com and that made it more popular… and therefore search engines ranked it higher, creating a feedback loop, where it eventually crushes the .net version.
Andrew Allemann says
And if you really want to see the future of search for competitive commerce terms, go google “ipad”
Dk says
I actually been tracking few .nyc domains and they are coming up better then a lot of of longer and more established domains on .com . At least for nyc searches. When i say a lot better, i means lot better. Some sites have only content and no links, and yet they are doing amazing. Hence google is bsing us.
Coupon pages, i dont agree that sounding will be main benefit. I truly belive .nyc will be values for SEO potential foremost, and branding and sound as strong secondary benefits.
couponpages says
“Coupon pages, i dont agree that sounding will be main benefit. I truly belive .nyc will be values for SEO potential foremost, and branding and sound as strong secondary benefits.”
It depends on the goals of your site. I agree that if a site gets the majority of it’s traffic from search engines, SEO potential will get a lot of first time visitors.
What I’ve noticed is that majority of the best ranked sites in the world have zero SEO value in their name, and even less of their traffic comes from search engines. What made them popular is radio / tv / print advertising, social media sharing, word of mouth and most important… return visitors, regardless of the original source.
For sites that basically sell products to people who are searching for them right then… search engines clearly are critical.
Al Gomez says
Thing is, TLDs play a major role in both SEO and online marketing: changing your domain name could prove to be confusing for both the algorithm and your users.
Anyway, a good domain name will always increase the chances of having returning visitors, which is why changing your current one (only if it’s bad) is a must. That being said, good call on the geo domains and local searches part – it’s easier to make a .co.uk site flourish in Google UK, as compared to others with a different domain name.
dereq777 says
all depends on how much seo you do on a site and the demand. all of it varies in time, location, site, quality, etc. there is really no definite answer for this. you can make a guess… but you will never know for sure.
Christopher Hofman Laursen says
Hi Andrew, some very valid points.
I agree that it’s all about relevance for Google, and less about the TLD. But some new gTLDs are already more relevant than their .com or ccTLD counterparts such as the city / regional TLDs. I assume also that some high barrier TLDs such as .luxury will also do well in general.
An interesting study to undertake would be how CTR friendly a new gTLD url is compared to .com. Globerunner already did a nice test for Google Adwords at http://globerunner.com/com-vs-new-gtld/
Steve B says
Don’t even go there. It’s just going to open up new doors for spammers to manipulate search engine rankings. And then, it’ll be EMD algo update part 2.
Josh says
Regardless of whether the new TLDs help or hurt as far as ranking goes, my opinion is that unless it is a well known brand using one of the new TLDs is that customers in general will be more leery than when shopping a .com