Copyright and your website: protecting your content and respecting others'

Copyright works in two directions, and a website owner sits on both sides of it. Original text and photos you create yourself are generally protected automatically. But work you commission — your logo, your site's code, a photographer's shots — is a trap: copyright normally starts with the human author who made it, so commissioning and paying for something does not, by itself, make you the copyright owner. At the same time, every image, font, and paragraph you place on your own site must be something you have the right to use. Getting one side right and the other wrong is how well-meaning site owners end up on the receiving end of an infringement complaint — or with a copycat they cannot stop.

This is practical guidance, not legal advice. Copyright rules and remedies vary from country to country and turn on the exact facts and contract, so for anything with money or reputation at stake, check your local law and talk to a lawyer.

Protecting your own content

Copyright is generally automatic: it arises without registration or a © symbol when a human author creates original expression that meets the applicable country's requirements. (Facts, ideas, and commonplace phrasing aren't protected as such, and some countries require the work to be fixed in concrete form.) Belgium and every EU member state participate in the Berne system, as do most countries you'll deal with — national law still decides the details. Automatic protection doesn't make registration pointless, though: depending on the country, a registration or deposit can strengthen evidence or affect remedies — and for US works, registration (or refusal) is generally required before an infringement lawsuit. What helps most in any dispute is evidence of authorship and date:

• Keep originals — RAW photo files, layered designs, draft documents with timestamps.
• Keep your version history. Git commits, CMS revision logs, and dated backups support your account of when a work existed — strongest when several independent records agree — though ordinary timestamps aren't conclusive proof of authorship or first publication.
• Public captures help too: an archived copy in the Wayback Machine is independent evidence the content was publicly available by the capture date — not of who wrote it or when it first appeared.

To find copies, paste a distinctive sentence from your site into Google in quotes — an exact-phrase search surfaces indexed pages using your words — and reverse-search key images (Google Lens, TinEye). Neither is comprehensive: unindexed, blocked, or modified copies slip through, so treat these as spot checks.

Before you complain, check your footing: confirm you own the relevant rights, that what was copied is protected expression rather than facts or generic wording, and that no quotation, criticism, parody, or other local exception plausibly covers the use. Then escalate in steps: a polite email first (many people genuinely don't realize), then a formal notice with a deadline. If that fails, check whether the host, platform, or search engine publishes a copyright or notice-and-action procedure and follow that service's current requirements — what it is called and how it works depends on the platform and the applicable law. Note too that offering the copier a link and credit is a possible settlement you might accept — it does not by itself cure the infringement, which needs the rightsholder's actual agreement.

Get commissioned work assigned to you in writing

Copyright usually begins with the person who created the work — the developer, the designer, the photographer — so don't assume paying transferred it. (The law adds exceptions: employee works and, in some countries, certain commissioned works can vest differently — which is exactly why the contract and governing law need checking.) Put it in writing before or when you commission:

• Get a written assignment of the economic rights, or at least a broad license wide enough for how you'll actually use the work — including future changes, so you're not locked in when you later edit or rebuild. Local formalities matter; a vague clause may transfer less than you think, and moral rights can stay with the author regardless.
• Have the agreement identify each deliverable — code, design, photography, illustration — and list third-party components separately: fonts, stock assets, themes, and open-source code stay under their own licenses, because the supplier usually cannot assign what isn't theirs. Check those licenses transfer to you or get your own.

A short assignment/license clause in the contract is far cheaper than discovering, after a falling-out, that you can't touch your own website.

Not infringing on others

The trap that catches most small sites: you cannot simply lift an image from Google Images. Search results are not a free library; a result page supplies no permission. Use assets whose use you can document — through ownership, a valid license, verified public-domain status, or an applicable exception:

• Licensed stock — free libraries (for example Unsplash or Pexels) and paid ones (for example Shutterstock or Adobe Stock) both exist; read each one's license (they differ and change) and keep a copy of the terms you agreed to. For commercial use, check whether recognizable people, private property, artwork, or brands need additional releases — a copyright license alone doesn't cover those.
• Creative Commons, but read the exact license and version: attribution (BY), non-commercial (NC), share-alike (SA — adaptations must carry the same license), and no-derivatives (ND) all change what you may do. Comply with the attribution format, and be aware the person who applied the license must actually have held the rights.
• Fonts — digital font files are licensed (and may be protected as software; protection of the typeface design itself varies by country). A desktop license doesn't necessarily permit web embedding, so check. Many open-source web fonts allow self-hosting (download and serve the files yourself) — prefer that, where the license permits, over hotlinking a third-party font service, which sends visitors' connection data (including IP addresses) to that service and has raised GDPR objections in the EU. Self-hosting settles that one request, not your site's privacy compliance overall.
• AI-generated images and text sit in a gray zone: in many jurisdictions purely machine-generated output may not be copyrightable, and generators can reproduce protected material. Don't assume "the AI made it" means it's clean.

Treat copyright as a two-way street: document and secure what you own, license what you use — and check local law when it matters.

What to do

1. Get a written assignment or broad license for all commissioned work — each deliverable named, future changes included, third-party components listed with their own licenses.
2. Keep dated originals and version history. A footer copyright line identifies the claimed owner and deters casual copying — it doesn't create or prove your rights — and have important site terms drafted appropriately for your jurisdiction.
3. Periodically search exact phrases and reverse-search key images to catch copycats, knowing the tools aren't exhaustive.
4. Check your rights and possible exceptions first, then escalate: polite email, formal notice, then the service's current copyright procedure.
5. Only publish assets whose use you can document — ownership, license, public domain, or exception — never "found it on Google".