AI

When to use AI to build websites, and when not to?

You might be thinking about using AI to build your entire website or some parts of it. This guide is for solo founders, freelancers, small business owners and beginner marketers trying to decide whether AI is the right way to build their site.

Here’s the short version

AI makes building a website a lot easier, but it can make maintaining it a lot harder. Building a website might seem like a gargantuan task if you’ve never done it before. But what’s harder than building a website is maintaining it over the long term, and the decisions you make at the start decide how easy or hard that will be.

What I mean by building with AI?

When I say building with AI, I mean using ChatGPT, Lovable or any other AI website builder to create your site, where you describe what you want and the tool generates the pages for you.

I’m not saying never use AI for anything on your website. If you want to add a section to highlight a product or make something look nicer, definitely use AI, as long as it doesn’t break anything else. But if you’re thinking of AI as the primary way to build your entire site, that’s something you need to think through carefully. I’m also only talking about websites here, not web apps.

Where building with AI works well?

  • AI is great for getting you off the ground. If you’re not a deeply technical person and you want to start a business, building your site with AI is a good way to test your idea. A 5-page website that might take a couple of days to build manually can be built with AI in less than an hour.
  • Landing pages for short-term campaigns. If a page runs for a few weeks to a few months and doesn’t need updates once it’s live, AI is a great fit.
  • Portfolio and personal sites that rarely change. If you need a site to show your work to potential clients or employers, or just to share information about yourself, AI can get you there quickly. But if you plan to add new work regularly or use the site as a personal blog, a platform like WordPress is the better choice. Imagine writing a blog post every week and asking AI to turn it into a page and push it to your site. Your posts need to be formatted well, your images need to be hosted properly and your pages need to be set up correctly to be found by search engines. Doing that through AI every week gets clumsy fast, and platforms like WordPress already solve these problems.
  • Small business websites with fewer than 10 pages.
  • Non-profit and community websites where the cost of building is a big factor.

When building an entire site with AI is a definite no!

  • E-commerce websites. Creating multiple product pages, even for a small number of products, and handling orders and payments is a complex process. Existing platforms like Shopify take care of all this properly.
  • Websites with more than 10 pages. That’s the threshold I’d set for myself when building an entire site with AI. Beyond that, adding and maintaining pages becomes complex.
  • Websites that need to be updated frequently. Content-heavy websites, like news sites or blogs with new content added every day.
  • Websites where design consistency matters. AI tools don’t maintain design consistency that well. They’ve gotten better over time, but they still struggle when you add a new page or a new section. The consistency slowly fades as you add more to your site. You can offset this a bit by keeping a strict design guidelines document that the AI tool refers to whenever it adds something, but in the end it depends on how consistent the tool itself is.

Why maintenance is the real test?

By maintenance, I mean the everyday work of running a site: adding new pages, adding new information and images, making changes to existing pages and keeping your content up to date.

Companies and developers have spent years building tools that solve these problems. Shopify made running an online store much easier. WordPress made it much simpler to build and manage content and business websites, while giving you control over your own data. For a new or less technical person, building with AI can seem like the simpler route, until maintaining what you’ve built becomes a chore.

Signs you’ve outgrown your AI-built website

  • Minor changes, like updating text on a page or adding an image, are taking longer.
  • Your website gets slower as you add more pages.
  • You keep having to rebuild things because the design feels inconsistent.
  • You need to scale up. You used an AI-built site to launch, and now things are real: you’re getting more traffic and more enquiries, and you need to add more information and optimise the site for a wider audience.

If maintaining your site is eating a big chunk of your time for simple things like these, it’s time to consider rebuilding it on a proper platform.

How to make the right decision

Do your research before choosing between AI and a dedicated platform. Think about what you need the site to do, how often you’ll need to update it and how much content it will eventually hold. AI websites are a great way to start something, test an idea and figure out how you want your site to look. Just go in knowing where the limits are.

Your website is a portal to your business, your work or your community. Treat it as an asset, not just another thing to have. A properly built and maintained website helps you get discovered and brings in new opportunities. Don’t ignore this part.

A simpler understanding of what WebMCP is, and whether it matters to you

WebMCP is a proposal for a new web capability that lets AI agents execute actions on a website directly. It’s still in the early stages and currently only works in Chrome.

What does this mean?

AI chatbots like ChatGPT can already go through a website in a browser, click things and perform actions. But it isn’t smooth and it doesn’t always work. While an agent is clicking through a site, a pop-up might appear, or the page might shift because an element hasn’t finished loading. Pages can also be long and heavy, which makes it harder for an agent to find what it needs.

WebMCP aims to fix that by giving agents clear instructions for a set of pre-defined actions like filling a form, adding a product to a cart, filtering through products, booking a ticket and so on. Instead of working out how your site behaves, the agent is told. That makes the action smoother and quicker.

How would this work?

Think of a simple contact form. Right now an AI agent fills it the way a person does. It finds the page, works out what each field is for, types into each one, then submits.

With WebMCP you add a few extra tags to the form that tell the browser what the form is for and what each field expects. The agent asks the user for exactly the information the form needs, then the browser jumps to your form and fills it in correctly, on screen where the user can see it. The user usually still clicks Submit, or the control of that can be given to the agent as well .

Plain HTML form

<form action="/submit">
  <label for="firstName">First Name</label>
  <input type="text" name="firstName" id="firstName">

  <label for="lastName">Last Name</label>
  <input type="text" name="lastName" id="lastName">

  <button type="submit">Submit</button>
</form>

HTML form with WebMCP

<form toolname="createSupportRequest"
      tooldescription="Submits a request for customer support."
      action="/submit">
  <label for="firstName">First Name</label>
  <input type="text" name="firstName" id="firstName">

  <label for="lastName">Last Name</label>
  <input type="text" name="lastName" id="lastName">

  <button type="submit">Submit</button>
</form>

For things that aren’t forms, like filtering hundred products by different sizes, colors, etc., you can write a small JavaScript function instead and give it a name and a description. The agent calls that function directly and gets clean results back.

This doesn’t mean agents will act the way you want them to

You can pre-define how a form on your website should be filled, but there’s no guarantee an agent will use it. Just as a person can do whatever they like on your site, so can an agent. It might ignore WebMCP entirely and click through the form fields instead.

You can tighten control over the actions that matter most, but you can’t force agents to take that route. And right now, most don’t. Browsers are only just adding support, and the major AI agents aren’t calling these tools yet.

Where is this really useful?

Website with long forms with a lot of fields, multi-step forms, hundreds of products or services behind multiple filters, complex booking/reservations processes, etc. These are all actions you can describe for an agent using WebMCP, and they’re the ones where an agent clicking its way through is most likely to go wrong.

Does this matter to you?

For most websites, not yet. If your site is a handful of pages where people read something and maybe send an enquiry, an agent can already handle that well enough, and there’s nothing here worth acting on today. The sites that should be paying attention are the ones where something actually gets completed on the page such as booking systems, e-commerce checkouts, large catalogues with filters, multi-step applications, customer portals and internal tools. The more steps between arriving and finishing, the more there is for an agent to get wrong, and the more WebMCP has to offer.

Even then, this is worth watching rather than building right now. The agents that would use it haven’t shipped support yet fully, so implementing today buys you readiness, not traffic. The useful thing to do in the meantime is to notice which parts of your site would be hardest for someone to complete without seeing the screen. Those are the parts you’d describe first when this becomes real.