Say a customer taps the Menu link on your restaurant site and lands on a page that says the page cannot be found. They do not email you to report it. They do not try again later. They go back to the search results and click the next place down the list. You never hear about it, and that is exactly the problem: broken links and 404 errors rarely announce themselves to the owner. They just quietly leak visitors while everything looks fine from where you are sitting.

A broken link is any link that points somewhere that no longer works. A 404 is the error a visitor sees when the address they asked for does not exist. Both can chip away at the trust a visitor places in your business, and both tend to build up slowly, which is why so many small business sites carry a few without the owner ever knowing. Here are the specific mistakes that let them pile up, and what to do about each one.

Hand-typed links that no one checks again

When you or whoever built your site typed out the address for a link, one small slip, a missing letter, an extra slash, a page name that later changed, is enough to send every click to a dead end. Nothing looks wrong on the page. The link is blue, it is underlined, it looks completely normal. It just goes nowhere.

The reason this hurts is that the visitor blames you, not a typo. A dead link on your own site can read as carelessness, and a visitor who was about to book or call may quietly decide you are not paying attention. To catch these, actually click your links rather than looking at them. Start with your main menu and footer, then follow the buttons that matter most: Book, Contact, Order, Menu. If a link goes where you expected, move on. If it does not, you have found one.

You rename or delete a page and leave the old address dead

This is the one owners cause without realising. You tidy up the site, rename the Services page, remove an old seasonal offer, or rebuild the whole thing. The new pages work perfectly. The trouble is that the old addresses may still be out there: in Google's results, in a printed flyer, in a customer's saved bookmark, in a post you shared months ago. Every one of those now points at nothing.

Imagine a plumber who deletes an old /emergency-callout page and builds a better one at a new address. A customer with a burst pipe taps a months-old link from a local listing, hits a 404, and calls a competitor instead. The page exists. The visitor just could not reach it.

The fix here is a redirect: a quiet instruction that tells the browser the old address has moved to the new one, so the visitor lands on the right page without ever seeing an error. You usually cannot set this up from the visible part of your site, which is why it tends to get skipped. When you change or remove a page, the question to ask whoever manages your site is simple: where does the old address now send people?

The default error page left as a dead end

Every site shows something when a visitor asks for a page that is not there. Left alone, that something is often a bare, unbranded message that offers no way forward. The visitor has shown they want your business, and the page answers with a shrug.

A good 404 page turns a dead end into a doorway. It should carry your normal header and menu, say plainly that the page could not be found, and offer clear links onward: your home page, your contact details, your main service. That way a visitor who hits a broken link has a reason to stay instead of leaving. Ask whoever maintains your site whether your error page is a designed part of the site or the plain fallback, and if it is the plain one, get it replaced with a real page.

You link out to other sites and assume they stay put

Links to your suppliers, your social profiles, a booking tool, a downloadable menu or price list: you control none of those. The other side can move a page, close an account, or take a file down, and your link to it breaks without a single change on your end. These are the broken links owners almost never think to recheck, because they worked perfectly the day they were added.

You cannot stop the other side from moving things, so the answer is to check the outbound links now and then, and to treat any button that takes money or bookings as the highest priority. If an external link matters to your revenue, it deserves a look on a schedule, not a look only when a customer complains.

After launch, nobody is watching the site

This is the mistake underneath all the others. A site is not a finished object like a printed sign. Links rot, pages move, files disappear, and the software behind the site changes over time. A site that was flawless at launch can accumulate broken links month after month, and if no one is responsible for noticing, no one does, until a customer or a competitor's better-kept site makes it obvious.

Here is a plain sweep you can run yourself, or hand to whoever looks after your site. It is named for what it checks, not for how long it takes.

The Broken Link Sweep, six checks:

  1. Every navigation and footer link. Click each one and confirm it lands where the label promises.
  2. Every money link. Follow every Book, Order, Buy, and Contact button to the end.
  3. The same, on a phone. A link can work on a desktop and misbehave on mobile, so repeat the important ones there.
  4. Outbound links. Test your social profiles, any downloadable files, and any button that points to an outside tool.
  5. Your error page. Type a made-up address on your own site, something like your address followed by /this-does-not-exist, and see whether the page that appears helps the visitor or abandons them.
  6. Anything you recently changed. If you renamed or removed a page, confirm the old address now redirects somewhere sensible.

Run through those and you can catch much of what a casual visitor would run into, and probably a few things they would not.

Broken links are a maintenance job, not a one-off task

Catching broken links is not a clever trick you do once. It is upkeep, the same as sweeping the shop floor, and it needs an owner. For a busy small business, that owner is rarely the person running the counter, because the sweep only works when someone remembers to do it and knows how to set up a redirect when a page moves. Whether that person is you with the checklist above, or a human team whose job is to keep the site healthy so you rarely have to think about it, the point is that someone is watching. A site nobody checks is a site quietly deciding, on your behalf, how many visitors reach the front door.

Related articles