How often should you update your website? Six signs yours is overdue

Most service business owners ask the same question in roughly the same way. Not "is my website any good" but "how often am I supposed to be doing something to it?"

It is a fair question, and the honest answer is unsatisfying: there is no schedule. Nobody should be redesigning a website every two years because a calendar said so. Plenty of sites go four or five years and are fine. Others need work eighteen months after launch because the business moved and the site did not.

What actually exists is a set of triggers. When one of them fires, the site has drifted from the business, and that gap is where enquiries quietly go missing.

Here are the six that matter, and a check you can run on each one this afternoon.

1. The site describes a business you no longer run

This is the most common one and the least noticed, because you are looking at your own website through the memory of writing it.

Open your services page. Read every service listed, and cross off anything you no longer offer, no longer want, or have not been asked about in twelve months. Then write down the work you actually want more of, and check whether it appears anywhere on the site.

Service businesses change direction faster than their websites do. You drop the low-margin jobs. You pick up a specialty. You stop travelling two hours for a callout. None of that reaches the site, so the site keeps confidently advertising the business you ran three years ago, and keeps bringing you the work you were trying to move away from.

We are not immune to this either. When we closed two of our own services this year, our website was not the first thing to catch up, and building websites is what we do for a living. Nobody is above the drift. The difference is whether you check for it.

2. Your navigation grew by accretion

Menus only ever get longer. A page gets added for a new service, another for a campaign, another because someone suggested it. Almost nothing is ever removed.

Open your site on a phone and count the taps from landing on the homepage to your most profitable service page. One tap is good. Two is acceptable. Three or more and the thing you most want to sell is buried beside a page nobody has opened since 2022.

While you are there, read your menu labels out loud. "Solutions" and "What We Do" and "Our Approach" are the three most common labels on service business websites, and none of them tell a visitor anything. Name the service. If you do bathroom renovations, the menu should say bathroom renovations.

3. It does not say where you work

Search your own site for your service areas. Not the contact page, which everyone remembers to update. The homepage, the service pages, the places a visitor actually lands from Google.

If someone cannot tell within a few seconds whether you cover their suburb, a good share of them will leave rather than fill in a form to ask. Service areas expand and contract as a business grows, and the website is usually the last thing to hear about it.

4. Something on it is visibly stale

Four things to check, and they take about a minute between them:

• The copyright year in the footer
• The date on your most recent blog post or news item
• Any page that says "coming soon"
• Any team member who left

None of these are individually fatal. Together they tell a visitor the same story: nobody is minding the shop. For a service business asking a stranger to let them into their home or trust them with a job, that is precisely the wrong impression to leave.

A "coming soon" page that has been coming soon for two years is worse than no page at all.

5. You have quietly stopped turning up in search

Rankings decay. Not because you did anything wrong, but because competitors keep publishing and search engines keep changing what they reward. A site that ranked well at launch can slide over two or three years without a single thing on it changing.

Open Google in an incognito tab and search the three or four phrases a customer would actually type. Service plus suburb. Note where you land.

Then do something most people have not tried yet: ask ChatGPT, or Google's AI Overview, who they would recommend for that job in your area. See whether you are named.

That second check matters more every month. A growing share of customers now get a shortlist from an AI assistant before they ever open a page of search results, and the businesses that get recommended are the ones whose sites are structured so a machine can read them properly. If you are invisible in both places, the website is not the asset you think it is.

6. You cannot make a small change yourself, so you do not

Pick one thing that is wrong on your site right now. A price, a phone number, a service description, a photo of a van you no longer own. Try to fix it.

If you cannot work out how, or you can but you have been putting it off for eight months, the site has stopped being a tool and become an obstacle. Every small change turns into an email, a wait, and possibly an invoice, so you stop making them. The site ossifies, and the drift in the other five signs accelerates.

That is a structural problem, not a discipline problem. And it is the single strongest argument for a rebuild rather than a patch.

So is it a refresh or a rebuild?

The useful way to decide is not by how old the site looks. It is by where the limit sits.

A refresh is the right call when the foundation is sound and the problem is on the surface. The structure holds up, it loads reasonably, you can edit it or we can. What it needs is current copy, better positioning, real photos, a navigation clean-out and a search visibility pass. You are improving something that works.

A rebuild is the right call when the foundation is the ceiling. If the site was built more than three or four years ago, the underlying platform is usually the limiting factor rather than the design. You can spend real money making an old build look better and still be stuck with the load times, the structure and the rigidity underneath. At that point a refresh is money spent on the wrong layer.

A rough rule that holds up well: if fixing the six signs above means changing what the site says, refresh it. If it means changing what the site can do, rebuild it.

The thing worth actually doing

You do not need a schedule. You need to run these six checks about twice a year, in about fifteen minutes, and act on whatever comes up.

Most businesses will find one or two things they can fix themselves in an afternoon. Some will find that the site has quietly stopped representing the business at all, which is worth knowing now rather than in another two years of enquiries that never arrived.

If you run the checks and are not sure which side of the refresh-and-rebuild line you fall on, that is a reasonable thing to get a second opinion on. It is a fifteen-minute conversation, and it is genuinely cheaper to have it before you spend money on the wrong layer.