Ask most people how to build a business website and the answer involves a content management system. It is the default, it has been the default for a long time, and for a great many sites it is heavier than the job requires.
A static site is the alternative. The pages are assembled once, when the site is built, and what the server holds afterwards is plain files. There is no database query when a visitor arrives, no template being rendered per request, no application code executing at all. A browser asks for a file and gets one.
That difference sounds like an implementation detail. It changes four things that businesses care about.
Speed, because there is nothing to do
A page that already exists is served faster than a page that has to be produced. No database round trip, no plugin chain, no template render on every visit. The same file can sit on a content delivery network close to the person asking for it.
This is not an aesthetic preference. Page speed affects search ranking, and it affects whether somebody on a mid-range phone on mobile data waits for you. The visitors you lose to a slow site never appear as a complaint — one of several ways an inexpensive site’s costs arrive later and get attributed to nothing.
A much smaller thing to attack
Most website compromises are not sophisticated. They are an out-of-date plugin with a known vulnerability, an admin login being tried against a common password, or an upload form that accepted something it should not have.
A static site has no admin login on the public internet, no plugin ecosystem, and no database sitting behind a form. There is very little to be clever against. What remains is the host and the build pipeline, both smaller and better defended than a public login page.
Hosting that costs almost nothing
Serving files is the cheapest thing a computer does. A static site runs happily on ordinary shared hosting or object storage, and it does not fall over when a post gets attention, because there is no application to exhaust.
Nothing rots while you are not looking
A CMS is software you now run, and running software means keeping it updated. A CMS installation left alone for a year is a liability that grows quietly. A static site left alone for a year is the same site it was, serving the same files, with no accumulated debt. This is the difference that matters most for small organisations.
If your organisation has nobody whose job includes applying updates, that is not a small consideration. It is the whole consideration, and who maintains this after launch is a question worth settling before anyone chooses a tool.
When a CMS genuinely wins
All of that said, the honest answer depends on who edits the site and how often.
A CMS earns its keep when non-technical people publish frequently and independently. If your marketing lead posts several times a week, needs to schedule things in advance, wants to reorder a page without asking anyone, and must not be blocked by a developer’s availability, a CMS is the right tool and a static workflow will be a source of friction every single week.
It also wins where you need the apparatus around publishing: multiple contributors, draft and approval steps, roles and permissions, a media library, an editor that shows the page as it is built.
And it wins where the content is genuinely dynamic — user accounts, personalised views, anything that must differ per visitor. A static site can bolt on services for search or forms, but if the core of the product changes per person, you are building an application, not a site.
The middle path most people miss
There is a third option that gets overlooked, and for a lot of business sites it is the correct one: a git-backed CMS.
The idea is straightforward. Your content lives as files in the repository. An editing interface sits on top of it, giving a non-technical publisher a normal form with fields and a preview. When they press publish, the tool commits to the repository, the commit triggers a build, and the site is rebuilt as static files and deployed — atomically, if you have set the deploy up so a partial upload cannot leave a mixture.
The publisher never sees git. The site never runs a database in production. You get version history on every content change for free, including the ability to see who changed what and to revert it, which no traditional CMS gives you without a plugin.
The trade-off is real but modest: publishing is not instant, because a build has to run. For a site that publishes a few times a week rather than a few times an hour, nobody notices.
How to decide in one question
Ask how often the content changes, and who changes it.
Changes rarely, and a developer is involved anyway — static, plainly. Changes often, by people who should not need a developer — a CMS, or a git-backed editor over a static build. Changes per visitor — you are building an application and this question was never the right one.
The mistake is not choosing either tool. It is choosing the heavier one by default, then discovering that the maintenance nobody was assigned is the maintenance nobody did.
We build fast, low-maintenance sites and set them up so someone non-technical can publish. Tell us who will be editing it.