RAPID

Published 29 July 2026 · 6 min

Site builder or code: run the numbers over three years

A builder is cheaper in month one and dearer by year three. Here is the arithmetic, where the line falls, and when a builder is genuinely the right call.

The builder versus code argument is usually fought on taste. Let us do the arithmetic instead, because it is simple and it ends the conversation faster than any opinion.

Year one

Builder: a subscription tier with a custom domain, a page assembler's time, a template. The client has a site within a week and pays noticeably less.

Custom build: two to three weeks of work at several times the price. Over year one the builder wins, and there is no point arguing.

What happens next

The subscription does not stop. It is paid every year, it grows with the exchange rate and with the platform's appetite, and you cannot cancel it without losing the site. A builder site cannot be moved: you take the text and the images and leave the build behind.

By year three the subscriptions typically catch up with the cost of a custom build. By year five they are comfortably ahead. And you still hold nothing you can carry elsewhere.

Speed and search

A builder ships its whole engine to every page, because it cannot know in advance which blocks you used. On fast office internet the difference is invisible. On mobile data on a bus it is very visible, and that is where most of your visitors are.

Search engines account for load speed. It is not the only factor and not the biggest, but all else being equal a slow site loses to a fast one.

When a builder is the right choice

  • You need to test an idea in a week and can throw it away
  • A three page business card site that will not change for years
  • A one off event: a conference, a promotion, an intake
  • The budget is below the cost of a build, and that is not negotiable

In all of those a builder saves money and time, and pushing a custom build would be selling you something you do not need.

When you need code

  • The site generates enquiries and speed affects how many
  • You plan to integrate 1C, a CRM or a payment provider
  • You need several languages with correct markup for search
  • The site will live longer than three years
  • It matters that the project belongs to you outright

We do both, but landing pages and corporate sites we hand code. Not on principle, but because over the lifespan those sites actually have, the arithmetic favours code.