You need to test an idea
A landing page and a payment link can validate demand before you spend on a brand. That is a good use of a builder.
Guide
Use an AI builder if you need a credible presence quickly, have a tiny budget and are willing to do the work. Hire a designer if the site has to earn enquiries, pass an accessibility review or be owned outright. Most businesses are in the second group; some are genuinely in the first.
The AI builder market is large and getting better. It is also sold on a promise that does not hold: that a generated site is equivalent to a designed one because both render HTML. The difference is not the markup. It is the decisions behind it — what the page is for, who reads it, what proof it needs, how it behaves on a slow phone and what happens when the business wants to leave.
This is not an argument against builders. A one-page site for a new tradesman, generated in an afternoon for £20 a month, is a rational purchase. The mistake is buying that and expecting it to compete with a site built around a specific buyer.
Comparison
| Criterion | AI builder | Bespoke build |
|---|---|---|
| Time to first version | Hours | Two to four weeks |
| Upfront cost | £0–£300 | £495–£3,000+ |
| Monthly cost | £15–£45+ | £15–£49 care plan |
| Ownership | Platform renders the site; export is partial | You own the code, domain and accounts |
| Performance ceiling | Often 1.5–3MB, script-heavy | Typically under 1MB, no third-party scripts |
| Accessibility | Generic output; issues common | Built to WCAG 2.2 AA principles |
| SEO control | Constrained by the platform | Full control of markup, URLs and schema |
| Custom functionality | Apps and plugins, with limits | Built to the workflow |
| Support | Chat queue | A named person |
Fit
A landing page and a payment link can validate demand before you spend on a brand. That is a good use of a builder.
A sole trader who needs a page and a phone number, and will not update it, does not need a custom build.
Builders reward effort. If you will write the copy, choose the images and maintain it, the economics can work.
A free or cheap site is better than no site. Start there, then upgrade when the enquiries justify it.
Limits
Cost
Illustrative, using typical figures. Your quotes will differ.
| AI builder | Bespoke Tier 2 | |
|---|---|---|
| Upfront | £200 | £995 |
| Monthly platform | £30 × 36 = £1,080 | — |
| Care plan | — | £20 × 36 = £720 |
| Apps and add-ons | £10 × 36 = £360 | — |
| Three-year total | £1,640 | £1,715 |
| What you own at the end | A subscription and a partial export | The code and accounts |
Decision
If the site is expected to generate more than a few thousand pounds of work a year, the ownership and performance difference usually justifies a bespoke build. If it is a placeholder while you establish the business, a builder is fine — and you can migrate later, keeping the domain and the content.
The worst outcome is not choosing the “wrong” option. It is paying agency prices for a template, or paying builder prices and expecting agency work. Ask what you are buying, what you own, and what happens on the day you want to leave.
Questions
Not inherently. The problems are the usual ones: thin content, slow pages, poor structure and no local proof. A well-structured builder site can rank. A badly structured bespoke site will not.
Sometimes, with effort. Content can usually be exported; design and functionality often cannot. Plan the migration before you have three years of content in a proprietary format.
If the site is a primary source of enquiries, yes. If it is a business card, a builder is probably enough. Be honest about which one you are buying.
It is already replacing some of the production work. It is not replacing the judgement about what a page should say, who it is for, or what happens when the business changes.
Tell us what you are trying to achieve. We will tell you honestly whether we can help, what it would take, and roughly what it would cost — before anyone signs anything.