Domain Names and SEO: What the Name Decides

What the name decides and what it does not

A domain name decides what people type, what they recognize, and — if you pick a country extension — which country Google reads the site as being aimed at. It does not decide what you rank for. Google's starter documentation lists keywords in a domain name among the things not to focus on, and says that from a ranking perspective they "alone have hardly any effect beyond appearing in breadcrumbs."

Three things about a domain are documented to matter: a country-code extension works as a geotargeting signal, a permanent redirect carries signals from an old domain to a new one, and a domain's previous use can hurt a site built on it later. Nearly everything else in circulation is undocumented or denied on the record — that an older registration ranks better, that a keyword string earns a boost, that one generic extension outranks another.

Exact-match domains: the 2012 update and the system still running

An exact-match domain, usually shortened to EMD, is a domain whose name is the literal text of a search someone might type: the query registered as a name rather than a name that identifies a business. In September 2012, Matt Cutts, then head of Google's webspam team, announced on Twitter a small algorithm change to reduce low-quality exact-match domains in results, describing it as affecting 0.6% of English-US queries to a noticeable degree. It was never written up as documentation.

The common claim today is that Google has gone quiet on EMDs since. It has not. Google's guide to its ranking systems lists an exact match domain system among those in current use, not among the retired ones, and describes it in one sentence: the ranking systems "consider the words in domain names as one of many factors to determine if content is relevant to a search," while the exact match domain system "works to ensure we don't give too much credit for content hosted under domains designed to exactly match particular queries."

Both halves are load-bearing. Words in a domain are an input to relevance, and a system exists to stop an exactly matching name from being worth more than the content under it. An EMD is neither a boost worth paying for nor a penalty worth fearing. What Google has never published is a definition of an exact match, a threshold, or any measure of how much credit the system withholds.

A keyword string is a weak name and an average ranking asset

The real decision is rarely about EMDs in the abstract. It is whether to pay a premium for a generic keyword domain rather than register a coined or brandable name. Beyond the ranking answer above, a generic string works against you three ways.

• It is hard to own. A descriptive phrase is weak as a mark, and a competitor can sit beside you with a near-identical name in another extension.
• It is not automatically your label in search results. Google generates the site name shown beside a result from the home page and from references to the site across the web, and says WebSite structured data is the most important input if you want to state a preference.
• It sets the editorial direction of everything you publish. A name built out of a query invites content built out of the same query, and filling pages with repeated keywords or lists of cities and regions is a documented spam policy violation rather than an optimization.

A keyword domain is not a mistake, and a name that is both descriptive and memorable is a good one. The error is treating the keyword as the reason to buy.

The extension: two different questions, two different answers

On generic extensions, Google's position has held since it published guidance on the new top-level domains in July 2015: keywords in a top-level domain give no advantage and no disadvantage in search, brand extensions carry no extra weight in crawling, indexing or ranking, and extensions that look regional, such as a city extension, are treated as generic in the same way .eu and .asia are. The current starter guide is blunter — the extension "only matters if you're targeting a specific country's users, and even then it's usually a low impact signal."

On country-code extensions, Google's international documentation says a country domain provides "a strong signal to both users and search engines that your site is explicitly intended for a certain country." That is a statement about audience, not ranking strength, and Google lists the cost in the same place: such a domain can target only one country.

Two qualifications. Google treats a set of country extensions as generic because users read them that way; its documented list includes .ai, .cc, .co, .fm, .io, .la, .me, .tv and .ws, and those carry no country signal. And the advice to take a generic domain and set a target country in Search Console is now stale: that report is deprecated, with Google stating country targeting there "was determined to have little value for the ecosystem, and is no longer supported."

Domain age is the most durable myth in the subject

No Google documentation describes domain age as a ranking factor: it appears nowhere in the guide to ranking systems and nowhere in the starter guide. Google staff have gone past silence. Asked in April 2017 whether domain age affects rankings, John Mueller of Google answered in one word: no. Asked again in March 2022, he wrote that domain age "helps nothing." Asked in December 2023 who benefits from the belief, he replied that it is primarily the people who want to sell you aged domains.

The related claim that a ten-year registration outranks a one-year registration traces to a patent application rather than any Google statement, and no Google documentation supports it. Register for several years to take the annual renewal off your list of risks, not for a ranking return.

The myth survives because the observation under it is real: old domains do frequently outrank new ones. An eight-year-old site has had eight years to earn links, publish, and build an audience that searches for it by name — all things Google does document. Registration age is a proxy for that history, not a cause of it, and buying the proxy does not buy the history.

What a domain's past actually carries

A domain's history does have one documented effect, and it runs opposite to the folklore. Google's spam policies define expired domain abuse as "where an expired domain name is purchased and repurposed primarily to manipulate search rankings by hosting content that provides little to no value to users," a policy added alongside the March 2024 core update. Its examples: affiliate content on a site previously used by a government agency, casino content on a former elementary school site.

The policy is narrower than the alarm around it. Buying an expired domain is not itself a violation; what it describes is an acquired domain plus content repurposed primarily to capture the ranking signals attached to its past.

Separately from policy, a prior state can slow you down. Answering a question in 2025 about a rebuilt site on a previously expired domain, Mueller said that "sometimes it just takes a lot of time for the old state of a domain to be shaken off (sometimes that's also the case when it was parked for a while), and the site to be treated like something new / independent," with nothing manual an owner can do to speed it up.

The due diligence that follows is not Google guidance, but it follows from what Google has published. Before buying an aged domain, check what it hosted through archived copies, check whether the inbound links and anchor text describe a plausible business, and be honest about whether you want the name or the links. If it is the links, you are paying a premium for the risk named in Google's spam policies.

Moving to a new domain without losing the traffic

Migration is where Google's documentation is most detailed, and where most damage is self-inflicted. The mechanic is permanent server-side redirects from every old URL to its equivalent new URL, and Google states that "301 and other permanent redirects don't cause a loss in PageRank." A permanent redirect is also what tells Google's indexing pipeline to treat the new URL as canonical; a temporary redirect keeps the old URL in results.

Timing is documented as ranges, not dates. Google says a small to medium site can take a few weeks for most pages to move, that larger sites take longer, and that temporary ranking fluctuation during a move is normal. Two durations get confused. The Change of Address tool in Search Console holds the association between old and new sites for 180 days, after which Google no longer recognizes any relationship between them; the separate recommendation is to keep the redirects "for as long as possible, generally at least 1 year" so all signals transfer. That tool is required when moving between domains or subdomains, and needs both properties verified at domain level.

The ways migrations lose traffic are mostly on Google's own list of mistakes:

• A staging noindex or robots.txt block left in place, removing the new site from results while the old one already redirects away.
• Redirects pointing at URLs that do not exist, because the mapping was never built page by page.
• Redirect chains. Googlebot follows up to ten hops, but Google advises keeping chains under five.
• Sitemaps still describing the old site, or a new one never submitted.
• Server capacity too thin for the recrawl, so the crawler meets errors as it reassesses every URL.
• Moving a large site in one night, when Google recommends testing one section first.

Subdomain or subfolder, and why the argument continues

Google's documented position is unenthusiastic. Its starter guide lists subdomains versus subdirectories among the things not to focus on, advising site owners to do whatever makes sense for the business. Mueller has said the same in office hours: in general Google sees these the same, and he would keep things together where possible, using a subdomain where the content really is different.

Many practitioners disagree, and it would be dishonest to pretend otherwise. Agencies regularly report traffic gains after moving a blog or help center from a subdomain to a subfolder. Those reports are not bad faith; the trouble is that such a move almost never happens alone. It arrives with a redesign, new internal linking, a content cleanup, or all three, and none of the published cases isolate the variable. Google says the two are treated the same, experienced people report otherwise, and no published test settles it.

The operational differences are not in dispute, and usually decide it. A subdomain is a separate host, which buys independent infrastructure and deployment but makes later consolidation a migration needing the Change of Address tool. Where a subdomain's home page carries no structured data, Google may fall back to the domain-level site name for it. Choose on those grounds, choose early, and treat changing your mind as a migration with every risk in the previous section.