A website that converts vs a website that looks nice

Every couple of months a business owner sends us a link to their site with some version of the same message: "the enquiries have dried up, we think we need a redesign." And more often than not, when we open it, the site looks fine. Genuinely fine. Nice type, decent photography, a hero animation someone was clearly proud of.

The problem is almost never how it looks. It's what the site is trying to do, and how confidently it does it.

Design for the next click, not the mood board

Every page on a converting site has an answer to a very boring question: what do I want this specific visitor to do next? On the home page it's usually "understand what we do and click into the most relevant service". On a service page it's "believe we can help and enquire". When you know the next click, the sections you need — and the ones you don't — become obvious. When you don't, you end up with three carousels and a testimonials block floating in space.

Hierarchy first, copy second

I see teams spend a week wordsmithing a hero paragraph that no one will read, because the real problem is that the hero is answering the wrong question. Before you polish a single sentence, decide the order:

1. What do you do, in a sentence a complete stranger will understand?
2. Who is it for?
3. What should I do next?

That's the fold. Everything underneath is either proof (case studies, testimonials, the odd credential) or specifics (services, pricing, how you work). Get the order right and the copy almost writes itself. Get it wrong and no amount of tone-of-voice work will rescue it.

Speed is a conversion feature, not an IT concern

We stopped treating page speed as a post-launch clean-up job about three years ago and it has made a bigger difference to enquiries than any single design decision. If your hero image is huge, your bounce rate is basically your image compression setting. On mobile, the drop-off gets steep past about three seconds — and that's before you even factor in how "slow" feels.

In plain terms: agree how big your images and videos can be before you start designing, not after. Keep chat widgets, analytics and marketing pixels on a short leash — they're usually the actual reason a site feels sluggish. A page that snaps open converts materially better than the same page that hangs for a moment. Same design, same copy.

Rebuilding or health-checking with Lovable and Claude Code

This is the part of the conversation that's changed most in the last two years. It used to be that a website rebuild meant a three-month project and a five-figure invoice before you saw anything real. That's no longer true. We now build and rebuild marketing sites with Lovable and Claude Code, with a senior designer and engineer directing the output, and the whole thing typically ships in a few weeks rather than a quarter.

The same tools make a website health check genuinely fast. If you already have a site and the enquiries have gone quiet, we can spend a few days looking at what's actually happening — how it loads on a real phone, whether the hero is answering the right question, where people are dropping out of the journey — and ship the fixes the same week instead of recommending a rebuild you don't need. Sometimes the honest answer is "your site is fine, change the form and add a proof section above it." Sometimes it's "the platform is holding you back, let's rebuild it in Lovable and move on with your life." Either is cheaper than the default of "let's redesign everything."

The other thing that quietly matters: whichever CMS or platform you land on, the best one is the one your marketing lead will open on a Tuesday afternoon without asking a developer for help. If content ships, the site works. If content sits in a Google Doc for three weeks waiting on someone with deploy access, the platform is the problem.

A short checklist we keep coming back to

• One clear promise above the fold. A visitor should know what you do and who it's for within about five seconds — in plain language, not jargon.
• One main action per page. Repeating the same button in a few places is fine. Two different buttons fighting each other is not.
• Proof close to the promise. Client logos, a quote, a real number if you have one — placed near the top, not buried at the bottom.
• Loads instantly on a phone. Under two seconds, no waiting for the hero image or fonts to pop in. This is a design decision, not a technical one.
• A contact form that asks for what you actually need. Nothing more. Every extra field measurably drops the number of people who finish it.
• You can see who actually enquired, not just who visited. Enquiries and bookings should show up in your analytics from day one, not "we'll set that up later".

Thinking about rebuilding? We do this with AI, so it's weeks not months.

If you're reading this because your current site isn't earning its keep and you're weighing up who to build the next one with — that's exactly what we do. Websites built with Lovable and Claude Code, with a senior designer and engineer on it, shipped in a few weeks instead of a quarter, with generous revisions included so we're not haggling over every change along the way. You get a site that's fast, clear, easy for your team to update, and — most importantly — actually built around getting visitors to enquire.

Have a look at how we approach website development, or get in touch and we'll take a proper look at your site with you — free of charge, and you'll get honest advice whether or not we end up working together.