Why AI makes good web developers more valuable.
Every few months, a client asks us “With all these AI tools now, do you still need developers to build a website?”
It’s a fair question. AI can generate layouts, write code, and produce content in seconds. After 26 years of building websites, we’ve never seen technology move this fast.
We were early adopters. We embraced AI tools as soon as they became viable, and they’ve genuinely revolutionised our productivity. But here’s what we’ve learned. AI has made skilled web developers more valuable, not less.
Here’s why.
AI is transformative for people who already have the skills
AI is genuinely empowering. It’s not hype. The productivity gains are real, and we’ve built them into how we work every day.
But those gains come from AI complementing existing expertise, not replacing it. Getting useful results requires knowing what to ask for, recognising when the output is wrong, and having the skills to refine and fix problems. Someone without web development experience won’t spot when generated code has security issues, performs poorly, breaks on certain devices, or doesn’t follow modern standards.
For us, we’ve found AI is a productivity multiplier. We can explore more ideas, prototype faster, and spend less time on repetitive tasks. But that’s because with our experience we already know what good looks like. The AI tools we use amplify that experience, they don’t substitute for it.
The gap between “something” and “the right thing”
AI excels at producing something quickly. Need a contact form? A hero section? Some placeholder copy? Done in seconds.
The challenge is that “something” often isn’t the same as “the right thing.” AI tools tend to generate generic solutions because they’re trained on common patterns. The output often looks plausible but lacks the specificity your business actually needs.
So it’s incredibly useful for rapid prototyping and exploring ideas., but somewhat less useful when you need a site that stands out, converts visitors, or integrates properly with your existing systems. That’s where human judgement and business understanding come in.
Strategy isn’t a prompt
A website isn’t a collection of pages, it’s a business tool. Before creating a design mock-up or writing a single line of code, we need to understand your brand objectives, your typical customers, your competitive positioning, and what success actually looks like for your business.
AI can’t sit in a meeting and hear that your sales team keeps getting the same three questions. It can’t notice that your competitors all look identical and spot an opportunity to differentiate. It can’t understand that your customers in Germany have different concerns than your customers in the UK.
This strategic layer determines whether a website generates leads or just takes up space on the internet. No amount of AI-generated code fixes a site built without this foundation.
The debugging problem
Debugging, i.e. finding and fixing problems, is where AI’s limitations become painfully apparent.
AI can be brilliant at suggesting solutions, but it can also hallucinate fixes that look plausible but don’t actually work. It can send you down rabbit holes, confidently insisting on an approach that’s fundamentally wrong. It can get stuck in loops, trying the same failed logic over and over with minor variations, unable to step back and question its assumptions. When you’re deep in a complex problem, an AI that keeps doubling down on a bad approach can waste hours.
Experience tells you when to ignore the AI and try something completely different. It tells you when the suggested fix is nonsense, even if it sounds convincing. It tells you when to stop, step back, and think about the problem from first principles rather than following a confident machine down yet another dead end.
When your WooCommerce checkout stops working late on a Friday afternoon, you need someone who can cut through the noise, diagnose whether it’s a plugin conflict, a server issue, a payment gateway problem, or something in your custom code—and actually fix it.
How AI has changed our work
We use AI tools throughout our process now. It’s transformed how we work to include:
- Rapidly generating and iterating on code
- Exploring multiple approaches to problems in a fraction of the time
- Accelerating documentation and routine tasks
- Getting instant answers to technical questions
- Reviewing code for potential issues
The productivity gains are substantial. Work that used to take days can sometimes be done in hours. But, and this is crucial, every output gets reviewed by experienced eyes. We know when AI has given us something brillian, but we can also recognise when it’s confidently produced something that won’t work. That judgement comes from decades of experience, not from the tools themselves.
The real question
The question isn’t whether AI can produce a website,it can, or at least it can generate something that technically qualifies. The question is whether that website will work for your business: loading quickly, ranking in search results, converting visitors, integrating with your systems, staying secure, and evolving as your needs change.
That requires understanding your business, making strategic decisions, solving novel problems, and taking responsibility for the outcome. These remain firmly human skills, but now turbocharged by better tools.
We’ve been building websites for over a quarter of a century and AI is the most exciting technology we’ve ever adopted. But the fundamentals of building websites haven’t changed.
That’s what we do. Now with better tools than ever.