There is a pattern we see often enough to predict. A business pays for a new site. It launches, and for a few weeks everyone is pleased. Then the designer's engagement ends, because that is what it was scoped to do, and the site enters a long quiet period where nothing is added and nothing is checked.
Two years later the owner says they need a new website. They usually do not. They need the eighteen months of small work that never happened, and a rebuild is simply the only way anyone has ever offered to sell it to them.
This is not a criticism of the owner. Nobody wants to manage a website. It is a genuinely tedious job that requires knowing a handful of unrelated things, and the entire point of hiring a company is not having to hold any of it. The problem is that the work gets sold as a project with an end date, and websites do not have one.
What actually goes stale, in order
It is rarely the design. Design ages slowly and a clean site from three years ago is usually fine. What goes first is everything factual, and it goes quietly:
- Hours, particularly after a holiday change that never got reverted
- Staff, including the person who left and the two who joined
- Services you have stopped offering and, worse, the ones you started and never added
- Prices, if any are published
- The contact form, which broke when something updated and now goes nowhere
- Photos of a location you no longer occupy
- A copyright year that tells every visitor exactly how long nobody has looked
Individually none of these cost you much. Together they do something worse than look untidy. They tell an attentive visitor that the business may not be paying attention generally, and the visitor who notices is often the careful one you most wanted.
The form nobody has tested since spring
If you take one thing from this, take this one. Contact forms break silently and they break more often than anyone believes. A plugin updates. A mail provider tightens something. The address the form points at belonged to someone who left. Nothing on the page changes. The form still says thank you.
The business finds out weeks later, usually by accident, and there is no way to recover the inquiries. They went to people who did answer.
The same applies to whatever else the site depends on. The phone number that is meant to be tappable on a phone. The booking link that points at a calendar somebody has since replaced. Anything that leaves your site and lands somewhere else is a thing that can break without telling you.
Most redesigns are deferred maintenance in a new coat of paint
When a site has been left alone for a few years, a rebuild starts to look like the sensible answer. It is not, usually, and it is an expensive way to discover that.
A rebuild throws away the things on the site that were working. Pages that had begun to rank, URLs other sites had linked to, content that answered a question well. Done carelessly it costs search visibility that took years to build, and the new site looks better while performing worse. We have been called in to fix exactly this more than once, and the repair is slower than the rebuild was.
Sometimes a rebuild genuinely is right. When the site cannot be edited without a developer, when it is slow in a way no tuning fixes, when it was never built to add pages and adding pages is the work, then yes. The point is that this should be a diagnosis, not a default.
What ongoing should actually cover
The phrase website maintenance has been devalued by people selling a monthly fee for an automated update and a backup nobody has ever restored. Here is what the arrangement should include if it is worth paying for:
- Someone checking that forms, phone links and booking flows actually work, on a schedule, not when you report it
- Platform and dependency updates, applied and verified rather than applied and hoped for
- Backups that have been restored at least once, so you know they work
- Factual changes made within days of you asking, without a quote for each one
- New pages as the business changes, because a site that never grows never ranks for anything new
- Speed watched over time, since sites get slower gradually and nobody notices the day it happened
- Someone reachable who already knows how your site is put together
That last item is most of the value and it is the hardest to buy back later. The cost of a stranger learning your site in an emergency is paid in the hours you can least afford.
A site that does not grow cannot rank for anything new
There is a search consequence to all of this that gets missed. A business that adds nothing to its website for two years is competing against businesses that added something every month, and the gap compounds in a way that is hard to close later.
This matters more now than it did, because a growing share of searches are answered without anyone clicking anything. An assistant summarising who does a particular thing in a particular city is drawing on pages that state things clearly. If your site has not said anything new since the launch, there is nothing new for it to find, and being quotable is not something you can catch up on in a quarter.
None of this requires a content mill. A handful of pages a year, each answering something customers genuinely ask, outperforms a burst of activity followed by silence.
Who should own it
Whoever owns it needs three things. Access that does not depend on a former employee's email address. Enough technical knowledge to tell a real problem from a warning that can wait. And a reason to look at it when nothing is wrong, which is the part that almost never survives an internal arrangement.
Internally that last condition fails reliably. The person nominated has another job, the site works most days, and checking something that is probably fine loses to everything urgent. It is not negligence. It is what happens to any task with no deadline attached.
So the honest options are to hire it out, or to put a recurring appointment in someone's calendar and treat it as real. Either works. Assuming it will happen without one of them does not.
The arrangement we prefer
We build sites and then keep running them, and we would rather do both than only the first. Partly because the site stays useful, and partly because the second year is where we learn which pages are doing the work, which is not knowable at launch.
If you have a site somebody else built, that is fine and common. We take those over, and the first thing we do is check the things that break quietly. If you want to know what state yours is in before deciding anything, ask us for a look. We will tell you if the answer is that it is in good shape and needs an hour a month.