The £40k Repaint: When a Website Refresh Is Just Expensive Decorating

The brief arrives on a Tuesday. It says: 'We're not looking for a big rebuild, just a refresh. It feels a bit dated.'

Ask what problem the refresh solves and you'll hear about a competitor's homepage. Ask which customers are failing to do what, and the room goes quiet. Ask how you'll know it worked in six months, and someone says 'traffic'.

That's not a brief. That's a mood.

'Refresh' is a budget word

Nobody says 'I'd like to spend forty thousand pounds decorating'. They say refresh, because refresh sounds reasonable. It sounds small. It sounds like something you can get signed off in a Tuesday meeting without a business case.

But a refresh is the most expensive thing you can buy badly, because it costs the same design and build hours as a strategic rebuild and delivers a fraction of the return. You still redesign the templates. You still migrate the content. You still QA the forms. You just do it without changing anything that actually affects whether people buy.

A repaint is a legitimate choice. It's just rarely the choice being made deliberately.

Decorating versus rebuilding

Here's the distinction that matters.

Decorating changes how the site looks. New typeface, softer palette, wider gutters, nicer photography. The structure, the offer and the journey stay exactly as they were.

Rebuilding changes how the site works. The navigation follows a decision your buyers actually make. The pricing page answers the three questions sales fields every week. The enquiry form asks for less and routes better. The measurement plan tells you which pages earn their keep.

If the reason people aren't converting is that your proposition is unclear, a new typeface will not fix it. It will make the unclear proposition look more contemporary, which is arguably worse — you've now spent money making a leaky funnel look premium.

Five questions before you sign anything

Run this diagnostic with the person who owns the number, not just the person who owns the brand.

One: can you name the page where you lose the most people? If the answer is a shrug, you don't have a design problem, you have a measurement problem. Fix that first — it's cheaper.

Two: has the offer changed in the last two years? New pricing, new segments, new service lines, a shift from project work to retainers. If yes, your site is probably describing a business you no longer are. That needs restructure, not restyling.

Three: is the problem message or mechanism? Message problems — unclear value, weak proof, wrong audience — are content and structure work. Mechanism problems — slow pages, broken forms, checkout friction — are technical. Neither is solved by a visual refresh.

Four: do you trust your analytics? If marketing and finance disagree on lead volume, stop. You cannot scope a rebuild on disputed numbers, and you cannot prove a refresh worked either.

Five: would fixing speed alone move the needle? Run Core Web Vitals on your top ten landing pages. If half are failing, a performance project may deliver more than a redesign — and it's a fraction of the cost.

Three or more 'no' answers means the refresh brief is hiding a real problem. Two or fewer means you might genuinely be in repaint territory.

The metrics that should set scope

Traffic is a vanity input. It tells you people arrived, not whether the site did its job. The numbers that should decide what you build are:

• Conversion rate by segment — because a blended figure hides the fact that one audience converts and another bounces.
• Qualified lead rate — enquiries that sales actually wants to speak to. This is the one your finance director cares about.
• Cost per qualified lead, split by channel, tracked over time.
• Revenue per session for ecommerce, or pipeline value per session for B2B.
• Task completion on service journeys — quote requests, claims, bookings, renewals.
• Organic entry-to-enquiry rate, which tells you whether the pages search sends people to are pulling their weight.

If your brief doesn't name two of those, it isn't a brief yet.

When a refresh really is enough

This isn't an argument against ever repainting. There are honest refresh cases:

Your parent brand has changed and you need to inherit the new identity quickly. You've got accessibility failures flagged in an audit and need remediation without a structural change. Your cookie banner and preference centre need rebuilding for UK GDPR before the next contract review. Your platform is reaching end-of-life and you need to move before it's forced. Your content has drifted and needs a pruning pass rather than a new architecture.

All valid. All scoped by a specific problem with a specific owner. None of them are 'it feels dated'.

What the repaint actually costs

The invoice is the small part. The bigger cost is eighteen months of deferred work: you've spent a quarter of internal goodwill, two rounds of stakeholder reviews and your one shot at a launch moment on something that doesn't move a single metric. When the numbers stay flat, the team concludes that digital doesn't work, and the real rebuild becomes twice as hard to fund.

The most expensive website project is the one you have to do twice.

Rewrite the brief

If you're handed a refresh brief this quarter, don't refuse it. Interrogate it. Turn it into three lines:

The problem we're solving is X. We believe Y will change it. We'll know within Z weeks because metric N will move.

If those three lines can't be written, you're not buying a refresh. You're buying a repaint, and someone will ask about the return eventually.