Website Redesign: When It Pays and How to Do It Right

Most of the redesign conversations I have start with some version of the same sentence: the website looks old. Sometimes that is true and it matters. More often the site looks fine and the real complaint is underneath. The phone does not ring from it, nobody on the team can change a headline without calling a developer, or a competitor launched something sharper and it stung.

I run Killerspots, and we build and rebuild sites for businesses across the country. The part of this work that never shows up in the design galleries is the part that decides whether the project was worth doing. A beautiful new site that drops half its Google traffic in the first month is a worse business asset than the ugly one it replaced. So this is the version of the redesign guide I wish more owners read before they signed anything: when to do it, when not to, and how to come out the other side with your rankings intact.

When does a website actually need a redesign?

Looks are the weakest reason to redesign, and I say that as someone who cares a lot about how things look. Visitors forgive a slightly dated design if the page loads fast, answers their question, and makes it obvious how to call you. They do not forgive a site that takes forever on a phone or hides the contact form three clicks deep.

The strongest reasons are functional. Google indexes the mobile version of your site, so a site that was designed for desktop and squeezed onto phones is being judged on its worst version. A site built on an old theme or page builder that nobody on staff can safely edit tends to go stale, because every small change becomes a ticket and a bill. And a business that has added services, dropped others, or moved into new markets often has a navigation that describes the company it was five years ago.

The last signal is the one owners underrate: the site is not connected to anything. Forms email a shared inbox that nobody watches, calls are not tracked, and there is no way to tell which pages produce customers. A redesign is the natural moment to fix that plumbing, and for many of the businesses we work with it is the part that pays for the project.

Is a refresh enough, or do you need a full rebuild?

This is the decision that saves the most money, and it rarely gets asked out loud. A refresh keeps your URLs, your platform, and most of your content, and changes what people see and how they move through it. It is faster, it carries almost no search risk, and it often delivers most of the improvement an owner was hoping for.

A rebuild makes sense when you are fighting the foundation. If the site sits on a platform that makes speed impossible, if every page is a tangle of plugins that break on update, or if the information architecture has to change because the business changed, patching the surface just delays the real project.

My test is simple. List the five things you most want the new site to do better. If most of them are about appearance and messaging, refresh. If most of them are about speed, editing, integrations, or structure, rebuild. Owners who run that list honestly often find they needed half the project they were about to buy.

How do you redesign a website without losing Google rankings?

This is where redesigns go wrong, and it is almost never the designer’s fault. Rankings belong to URLs and to the content on them. When a new site launches with a different URL structure and no redirects, every old address that Google had ranked starts returning an error. The authority those pages built over years does not transfer on its own, and the traffic falls off a cliff that looks mysterious to everyone who did not plan for it.

The fix is unglamorous and it has to happen first. Export at least twelve months of page data from Google Search Console and list every URL with clicks or meaningful impressions. Add the pages that earn links from other sites and the pages your ads or email campaigns point to. That list is the set of pages the new site is obligated to protect.

Then build the redirect map, one old URL to one new URL, each pointing at the closest equivalent page rather than dumping everything on the homepage. Google treats a pile of redirects to the homepage much like missing pages, so the specific match matters. Where you can keep a URL exactly as it was, keep it. The cheapest redirect is the one you never needed.

Two more things protect you. First, do not gut the content on pages that rank. A redesign is a tempting moment to trim copy into something sleeker, and trimming the text Google was ranking is a reliable way to lose the ranking. Tighten it, restructure it, but keep the substance. If the copy genuinely needs rework, website content writing that is built around what already earns is a different job from writing fresh pages for a brochure.

Second, keep the staging site out of the index. Development copies of a site get discovered and crawled more often than people expect. We caught our own preview builds recording visits in analytics this month and locked them down with noindex and production-only tracking. It is a small thing, and it is exactly the kind of small thing that creates duplicate content if nobody checks.

What should be settled before anyone opens a design tool?

The design phase is the fun part, which is why it tends to start too early. When a project jumps straight to mockups, the structure of the site gets decided by whatever the homepage comp implied, and the hard questions surface late, when changing the answer means redoing work.

Start with what the site is for. For most service businesses the honest answer is that it exists to make the phone ring and the form fill. Write down which actions count and how you will measure them. Then decide the page structure from what you sell and what people search for, using the Search Console inventory as the floor rather than a blank sheet.

Then settle where a lead goes after someone submits it. A form that sends an email to an inbox is the lowest version of this. A form that drops the contact into a CRM, triggers an immediate follow-up, and tells you which page it came from is the version that pays. We wire new builds into LeadConnector for exactly that reason, because a site that captures leads nobody follows up on has only solved half the problem.

What does a new website need for AI search now?

A few years ago this section would not have existed. Now it is one of the strongest arguments for rebuilding rather than refreshing, because a lot of older sites are built in ways machines struggle to read. Service descriptions baked into graphics, important details hidden behind scripts that only load on interaction, and pages with a headline and a stock photo but almost no text all give an AI system very little to quote.

The sites that get cited tend to share a shape. Pages lead with a clear answer to a real question and then support it. Headings describe what the section actually says. The business name, services, locations, and contact details are consistent everywhere they appear. And structured data does the translating, which is why schema markup for AI search belongs in the build specification from day one instead of being bolted on later.

None of this replaces traditional SEO. It is the same foundation with more emphasis on clarity. If you want the broader picture of how the two fit together, our guide to AI search optimization covers the rest.

What goes wrong most often on launch day?

I keep a short launch checklist because the failures repeat across projects, regardless of who built the site. The worst one is the noindex tag. Staging sites are correctly told not to appear in search, and when that setting ships to production the new site quietly asks Google to remove it. Nothing looks broken to a visitor, which is why it can go unnoticed for weeks.

Right behind it are forms. Submit every form on the live site and confirm the lead arrives where it should. Call the tracked numbers. Check that analytics and any ad conversion tags are firing on the new pages. Then test a sample of old URLs from the redirect map and confirm each lands on the intended page rather than an error.

After that, watch Search Console closely for the first several weeks. Coverage errors, a spike in pages not found, and sudden drops on specific queries tell you exactly which part of the plan slipped, while it is still easy to fix.

How long does a website redesign take?

Owners usually assume the build is the slow part. It rarely is. The slow parts are gathering content, getting decisions made, and waiting on approvals from people who have day jobs. A project where the business has its service descriptions, photos, and a decision-maker ready moves quickly. A project where every page waits on a new draft and a round of comments does not.

The practical way to shorten the timeline is to make the content decisions early. Settle the page list and the inventory before design begins, gather real photography of your people and your work, and name one person who can approve. Those three decisions do more for a launch date than any development shortcut.

When should you hire an agency for a redesign?

If your site gets almost no search traffic and you mostly send people to it from your business card, a capable template and an afternoon of careful work can take you a long way. There is nothing to protect, so the risk is low.

The calculation changes once the site is producing. At that point a redesign is a migration as much as a design project, and the person running it needs to understand redirects, content inventories, structured data, and lead routing as well as layout. That combination is what we put into web design services at Killerspots: design that converts, built on a plan that keeps what the old site earned. For multi-page company sites that usually means a business website build, and for campaign traffic a dedicated landing page often does more than touching the whole site.

We work with businesses nationwide, and we do the same work for companies in our home market through our Cincinnati web design team. If you are weighing a redesign and want a straight read on whether you need a refresh or a rebuild, send us the site. Sometimes the honest answer is that a few targeted fixes will get you most of the way there, and that is a good answer to get before you spend on a rebuild.