When a Website Redesign Is Worth It (and When It Isn’t)

Most businesses buy a website redesign when what they needed was a rewrite of one screen. The instinct is understandable, because a dated site is visible and embarrassing in a way that a buried phone number is not. But "it looks old" and "it does not produce work" are different problems with different price tags, and only one of them needs a redesign.

Key Takeaways

• Looking dated is rarely the actual problem. The problem is usually what the first screen says and how easy you are to contact.
• Five questions separate a rebuild from a repair, and most sites fail on repairs.
• Rebuild for structural reasons only: you cannot edit it, it is slow, it is unsupported, or it cannot be found.
• A redesign that keeps the same words usually keeps the same results. Content is the lever, not layout.
• If you do rebuild, protect what already works. URLs, rankings and tracking are the things quietly lost in a redesign.

Why the redesign instinct fires

A website ages in public. You look at it, you notice the styling belongs to a previous decade, and it feels like the thing holding you back. Meanwhile nobody ever looks at their own site and thinks "the phone number takes eleven seconds to find on mobile," even though that is far more likely to be what is costing them.

So the redesign gets bought because it is visible and it comes with a proposal. The first-screen fix does not get bought because it is cheap, unglamorous, and nobody sells it to you.

Then the new site launches, it genuinely looks better, and the phone does not change. That outcome is so common it should be the default expectation whenever a redesign brief is about appearance. If the words did not change and the structure did not change, results were never going to.

Our contractor website design guide covers what the first screen has to carry. Start with a free AI Visibility Audit if you want an outside read before spending anything.

The five questions

Run these in order. They are ordered by how much they cost you when the answer is wrong.

One. Can a visitor tell what you do and where you do it, in words, without scrolling? Not implied by a photo or a tagline. Stated. If not, that is a text change, not a redesign.

Two. Is the phone number tappable on a phone, above the fold? If it takes more than about five seconds to find and press, you are losing calls today. Also a repair.

Three. Can you edit the site yourself? If changing a sentence requires emailing a developer and waiting three days, the site will never improve, because nobody will bother. This is the first question whose "no" genuinely justifies a rebuild.

Four. Does it load quickly on a phone on mobile data? Not on your office wifi. Slow enough that people leave is structural and often not worth patching.

Five. Is it built on something still supported? An abandoned platform, a discontinued builder, or a plugin stack nobody maintains is a rebuild whether you like it or not, because it becomes a security problem before it becomes a marketing one.

Failures on one and two are repairs. Failures on three, four or five are the real reasons to rebuild. If you fail only one and two, a redesign is the most expensive way available to not fix your problem. Book a call if you want a straight answer on which bucket you are in.

What a repair actually looks like

Rewrite the first screen so it says what you do, who for, and where, in plain words. Make the phone number a tap-to-call link and put it where a thumb rests. Cut the contact form to three fields. Add a next step to the bottom of each page. Replace stock photography with real photos of real work.

That list is usually a week, sometimes a day. It is also the list that changes results, because it changes what the site says and what it asks for rather than how it is decorated.

Here is the uncomfortable arithmetic that makes the case. Platform data from one call-analytics provider covering home services puts the unanswered rate at 27% of inbound calls, with the average missed call worth roughly $1,200 in lost revenue and fewer than 3% of voicemail-routed callers leaving a message. PATLive reports 85% of people whose calls go unanswered never call back. If a buried phone number is costing you even one call a month, the repair pays for itself immediately and the redesign has not started yet.

Do the repairs first regardless. If you later rebuild, you will carry the improved words across, and you will brief the rebuild from evidence rather than from taste. See what our website work covers.

If you do rebuild, protect what already works

A redesign is the most common way businesses accidentally lose search visibility, and it happens through carelessness rather than bad design.

Keep your URLs, or redirect them properly. If a page currently sits at /services/panel-upgrades and the new site puts it at /what-we-do/electrical, every link and ranking pointing at the old address needs a permanent redirect to the new one. Skipping this is the single most expensive redesign mistake.

Keep the words that were working. Pages that already rank do so because of their content. A rebuild that replaces four hundred words of specific, useful text with a headline and a photograph will lose the ranking that text earned.

Keep tracking alive. Analytics and conversion tracking are usually installed on the old site and forgotten. If they do not make it across, you lose the ability to tell whether the redesign helped, which is how a redesign becomes an article of faith.

Check it before it goes live, on a phone. Not on a desktop in a meeting.

Google's helpful content guidance is worth reading before a rebuild brief is signed, because it describes what makes a page worth ranking in the first place, and it is mostly about substance rather than presentation.

The audience your old site already has that your redesign might lose

Something changed recently that most redesign conversations have not caught up with.

45% of consumers now use AI for local recommendations, up from 6%, per BrightLocal's 2026 survey of 1,002 US adults. In 177 observed ChatGPT local-service sessions published on Search Engine Land in January 2026, only 39.2% ended in a click to any business website, with users examining an average of 3.7 businesses first.

That means a meaningful share of your buyers are reading a description of your business rather than your design. Assistants build that description from the text on your pages. A redesign that replaces text with imagery makes you harder to describe, and you will never see the traffic you did not get.

Google's AI features guidance is direct about what this takes: no additional requirements beyond being indexed and eligible for a snippet, no special optimisations, no new machine-readable or AI text files, and keep important content in textual form. That last instruction is the one a visual redesign most often breaks. Adding Google's LocalBusiness structured data makes the core facts explicit, with the requirement that markup match visible text. Ask us to check what an assistant currently says about you.

Frequently asked questions

How do I know if I need a website redesign?

Run the five questions. If you can edit the site, it loads quickly, and it is on a supported platform, you almost certainly need a repair rather than a redesign, and the repair is the part that changes results.

How often should a website be redesigned?

Less often than most agencies suggest. A site with clear text, a working phone link and current proof can run for years. Rebuild when something structural breaks, not on a schedule.

Will a redesign improve my Google rankings?

Not by itself, and it can hurt them. Rankings follow content and links. If a rebuild changes URLs without redirects or replaces substantive text with imagery, visibility usually falls, and it can take months to recover.

What should I do before starting a redesign?

Fix the first screen, the phone link and the form on your current site. You will learn what actually moves your numbers, and you will brief the rebuild from evidence instead of opinion.

My site looks dated. Is that reason enough?

On its own, no. Fix the words and contact path first and give it a fair run. If the phone changes, you have your answer cheaply. Ask us for a second opinion before you commit to a project.

The honest test

Before you sign a redesign proposal, write down what you expect it to change and how you will know. If the answer is "it will look more professional," that is a real goal but it is not a lead goal, and it should be priced as such. If the answer is "more calls," then the brief should be about words, structure and speed, and most of that work does not require a new website at all. GetLocalLeads.AI's website design service starts from that question rather than from a template.