What Is a Domain?

In the most common sense, a domain is simply a website's name: zds.es, google.com, wikipedia.org. It's what someone types into a browser's address bar or reads on a business card next to a company name. Registering a domain means paying a registrar, such as Namecheap or a hosting provider's own panel, for the right to use that name for a year or more. Nobody owns a domain outright: it gets renewed periodically, and if the renewal lapses, the name becomes available for someone else to register. This is the meaning almost everyone has in mind when they ask, "what's your domain?"

Underneath that name sits a hierarchical technical structure that belongs to the Domain Name System, DNS: the distributed database that translates readable names into IP addresses so browsers know which server to connect to. A domain like zds.es breaks down into a top-level domain (.es), a second-level domain (zds), and optionally one or more subdomains added in front, such as blog.zds.es or de.zds.es. The next section walks through this structure piece by piece, since it's the foundation for everything else in this article.

In an SEO context, domain picks up a third meaning: it's the unit that authority and link metrics get calculated on. When a tool like Ahrefs or Moz talks about referring domains, Domain Rating, or Domain Authority, it isn't counting individual pages or URLs, it's counting whole domains. That's exactly why the term shows up so often in any SEO glossary or report, not as a synonym for "website," but as the measurement unit that bundles authority, trust, and inbound links for an entire site.

Mixing up these three meanings causes real confusion in day-to-day agency or in-house work. A client who asks "is my domain good?" might mean whether the name is memorable and easy to type, whether the DNS setup is configured correctly, or whether their Domain Rating is strong against direct competitors. Each question needs a different answer and different data, so separating the three layers before answering avoids vague or flat-out wrong statements, especially in a first meeting with a new client.

A domain reads right to left and is made up of labels separated by dots. In zds.es, ".es" is the top-level domain, or TLD: the rightmost part, indicating the category or country the registration belongs to. "zds" is the second-level domain, or SLD: the actual name a business or person chooses and registers. Together, zds.es forms what's known as the root domain, or registered domain, the smallest unit that can be registered and paid for through an accredited registrar.

TLDs split into several families. Generic TLDs, gTLDs, like .com, .org, or .net, don't require any special eligibility and have been among the most widely used worldwide for decades. Country-code TLDs, ccTLDs, like .es, .de, or .fr, identify a specific region, and some of them, like .es, carry extra registration rules set by the relevant national body. Since 2012 there's also been a wave of new thematic gTLDs, such as .shop, .digital, or .agency, meant to add business context to a name instead of relying on whether a .com is still available.

A subdomain is an extra label added in front of the second-level domain, separated by a dot: blog.zds.es, de.zds.es, shop.zds.es. Technically it's created with a DNS record, usually type A or CNAME, that the domain owner sets up freely through their management panel, without registering anything new or paying a registrar again. That makes it possible to create as many subdomains as needed, at essentially no cost, and to switch them on or off within minutes.

This is the nuance that trips people up the most: a subdomain is not an independent domain, it's a subdivision within the same registered domain. blog.zds.es still belongs to zds.es, sharing the same registration, the same owner, and usually the same DNS zone. That's different from a brand new domain, say blogzds.es, which would require its own registration, its own payment, and would carry no automatic technical relationship to zds.es, even if the name looks similar.

In practice, businesses use subdomains to separate environments or functions without leaving their main domain: app.zds.es for an application, api.zds.es for a programming interface, es.zds.es or de.zds.es for language versions. All of that still lives under the umbrella of zds.es, even though each subdomain can point to a different server, or even a different hosting provider, than the main site.

SEO tools calculate authority metrics like Domain Authority or Domain Rating at the domain level, not the page level, because most of the trust a link profile carries gets attributed to the site as a whole rather than to a single isolated URL. When many different websites link to several pages of the same domain, that link equity adds up for the domain overall; a brand new page with no links of its own can still benefit from that accumulated reputation through internal linking. Measuring only at the URL level would waste that aggregate signal and give an incomplete picture of a site's real strength.

That's why metrics like referring domains count unique linking domains, not individual links. If the same website links ten times to a given domain, those ten links count as a single referring domain: what matters is the diversity of external sources, not the raw link volume, because ten links from one website carry far less weight than ten links from ten different websites. It's a way of keeping a link profile artificially inflated by a single source from looking stronger than it actually is.

Domain and URL aren't the same thing, even though they're used interchangeably in everyday conversation. A URL includes the protocol, the domain, the path, and sometimes extra parameters: in https://zds.es/blog/article?utm_source=newsletter, the domain is only "zds.es"; everything after that, /blog/article and the parameter, is part of the URL but not the domain. A single domain can have thousands of distinct URLs, each with its own content, while all of them share the same root domain and inherit part of its accumulated reputation.

Domain and hostname aren't exact synonyms either. Hostname is the full address of a server within a network, and it includes any existing subdomain: in blog.zds.es, "blog.zds.es" is the complete hostname, while "zds.es" is the registered domain that hostname belongs to. Put another way, any registered domain can also function as a hostname on its own if there's no subdomain in front of it, but a hostname with a subdomain isn't a separate registered domain, it still belongs to the same registration and the same owner.

This distinction between domain, subdomain, and hostname has a practical consequence that gets debated a lot in SEO: the difference between publishing content on a subdomain versus a subfolder of the main domain. Google has stated on several occasions, including 2018 Office Hours documented by Search Engine Journal, that it treats subdomains and subfolders the same for ranking purposes, saying internally it sees "these the same." Third-party tools like Ahrefs or Moz don't always follow that same logic, though: for them, a subdomain can generate its own set of referring domains and its own Domain Rating, separate from the main domain, because they technically track it as a distinct entity within their link index. That gap between how Google treats it and how a third-party tool measures it explains a lot of the confusion over whether blog.mydomain.com or mydomain.com/blog makes more sense for a new project.

• Choose between a subdomain and a subfolder based on the actual goal of the content, not just technical convenience, and keep in mind that third-party tools may measure them as separate entities.
• Verify both the www and non-www version of the domain in Google Search Console, and consolidate whichever version isn't used with a 301 redirect.
• Renew the domain well ahead of its expiration date and enable a transfer lock so it doesn't lapse by accident.
• Periodically review which subdomains actually exist and whether they still serve a purpose, instead of letting forgotten test subdomains pile up.
• When evaluating a domain for a link purchase or an acquisition, look at the whole root domain, not just the specific page of interest.
• Register the TLD variants relevant to the brand, like .com and .es, and protect them against typosquatting.

• Confusing a subdomain with a subfolder when distributing a site's authority, assuming blog.domain.com automatically inherits all of domain.com's Domain Rating.
• Treating domain and URL as synonyms when talking to a client or writing a report.
• Assuming hostname and domain are exactly the same thing, ignoring that a hostname can include a subdomain.
• Buying an expired domain purely for its high Domain Authority or Domain Rating, without checking its history for parking pages, past spam content, or penalties.
• Letting a domain lapse by missing a renewal, allowing a third party to register it.
• Multiplying subdomains without a strategy, fragmenting links and signals that could otherwise consolidate under a single domain.