- Home
- About
- Service
- News & Insight
- Contact
- Resources
Innovative Thinking
-
Innovative Case
-
A business website usually becomes expensive to maintain long before anyone decides to replace it. The cost arrives as website technical debt: plugins added and never removed, customisations made in a place the next update will overwrite, and upgrades postponed because the last one broke the contact form. None of it shows on the page. All of it shows in the invoice, in how long a small change takes, and in how reluctant everyone has become to touch anything.
The seven signs below appear in the bill rather than in the browser. Three or more means the site is charging you interest, and the answer is usually a revamp of the parts carrying the debt rather than a rebuild of everything.
The term comes from software. Ward Cunningham coined the metaphor, and Martin Fowler’s summary puts the cost plainly: the extra effort it takes to add new features is the interest paid on the debt. On a business website the debt is anything that makes the next change harder, slower or riskier than it needs to be — and the interest is paid in developer hours you did not plan for.
It is not the same thing as maintenance. Maintenance is the work you schedule; debt is what accumulates when that work is deferred, or when the plan covers plugin updates and nothing else. We have written separately on what a maintenance plan should actually include. This is about recognising the bill in a site that never had one.
Some debt is a deliberate and sensible choice: a shortcut taken to open an online shop before a festive season is fine, provided somebody wrote down what was skipped. Debt you chose and recorded is a decision. Debt nobody remembers taking is a surprise invoice.


The clearest sign of all. Someone updated a plugin, the enquiry form quietly stopped sending, and updates now happen only when they cannot be avoided. Every month of deferral widens the gap between what is installed and what is current, and it is that gap which turns a ten-minute update into an afternoon of untangling.
Adding a service, swapping a photograph on the homepage, correcting the opening hours in the footer. If any of those needs a quote, the site was built in a way that locks its owner out. The bill is the smaller half of that problem: the site also drifts out of date between paid jobs, because nobody raises a purchase order to fix a typo.
Twelve plugins doing nine jobs. One WhatsApp button was replaced by a second WhatsApp button and the first was never deleted, one has been deactivated for a year, and one was abandoned by its developer three versions of WordPress ago. Each is code somebody else wrote, with its own update cycle and its own licence renewal. Count the active plugins on your site, then count how many a visitor would notice if they disappeared tonight.
A change made directly in a theme file rather than in a child theme is erased by the next theme update and paid for again. If the phrase “reapplied the fix” appears anywhere in your developer notes, that is interest, not principal.
No staging copy means every change is tested by customers. It also means nobody dares attempt the update that would clear three of the other signs on this list, so the caution compounds along with the debt. Staging costs little on most Malaysian shared hosting plans, and going without it is rarely a saving.
WordPress recommends PHP 8.3 or newer and will still run on 7.4, while warning that end-of-life versions expose a site to vulnerabilities. On the PHP side, only the 8.2, 8.3, 8.4 and 8.5 branches still receive security fixes, and 8.2 loses even those at the end of December 2026. If your hosting control panel still shows 7.4 or 8.0, that is not a preference left alone — it is a deadline that passed. The same applies to a premium theme whose licence lapsed.
Take twelve months of website spend and sort every line into two columns: kept it working, and made it better. A site in reasonable health leans towards the second. When almost everything sits in the first column, the website is absorbing budget that was meant to grow the business.

A site left completely alone still gets more expensive, because the ecosystem around it keeps moving. Patchstack’s State of WordPress Security in 2026 counted 11,334 new vulnerabilities across the WordPress ecosystem during 2025, 42% more than the year before. Ninety-one per cent of them were in plugins and nine per cent in themes. WordPress core accounted for six, all low priority.
Two figures there matter more than the headline. Just under half, 46%, were not fixed by their developer before the vulnerability became public, so updating promptly is necessary but not sufficient: a plugin whose author has gone quiet is a liability however diligent you are. And 1,966 were high severity, more than the previous two years combined.
Core itself is well looked after: a security team of more than fifty people reviews and patches WordPress, and critical fixes reach older versions automatically. The exposure sits in the extensions somebody chose, which is another way of saying it sits in sign three.
None of this argues for starting again. On most business websites the debt sits in a handful of places, and a revamp that keeps the platform, the URLs and the content that already earn is cheaper than a rebuild that discards all three.
If several signs describe your site, the wider diagnostic of when a refresh is no longer enough is the sensible next read, and the revamp or rebuild decision covers the one case where the debt is not worth repaying: a platform that cannot be brought back to a supported state at all.
Every working site carries some. The useful test is not how much exists but whether anyone can name it. If the person responsible can list what is installed, what breaks and what was skipped, the debt is managed. If nobody can, it is compounding.
Compare twelve months of upkeep against what a scoped revamp would cost, and include the changes you wanted but never requested because they were too expensive to ask for. Patching wins when the site is sound and one or two things are wrong. Once three or more signs apply, patching usually costs more over two years and leaves you with the same site.
Yes, because “working” describes today only. Vulnerabilities are published against versions rather than against sites, so an unpatched plugin becomes a known way in the moment it is disclosed. Turn on automatic updates for the plugins you trust, keep a staging copy for the ones you do not, and delete anything you cannot justify keeping.
We are usually called in when the invoices stopped making sense, not when the site stopped working. The first thing we produce is the list described above: what is installed, what it costs, what breaks, and how much is worth carrying forward. Our web development and design work then repays the debt in the order that costs the business most, keeping the URLs and the content that already earn their place. If this pattern sounds like your website, tell us what the last twelve months of website spend went on.