Your site is three or four years old. It looked fine when it launched. Now it feels dated, the phone photos load slowly, you can't edit half the pages, and somebody on your team keeps saying "we need a new website." But nobody can say exactly what's broken. So the project stalls for another six months.
That's the moment this website redesign checklist is for. Because "we need a new website" usually means one of two very different things. Either the site works and looks tired, which is a refresh. Or the site is fighting you every time you try to do business through it, which is a rebuild. One of those costs a few thousand dollars and takes a few weeks. The other costs more and takes longer. Picking the wrong one is how people spend $15,000 and end up with the same problems in a prettier wrapper.
What a refresh actually fixes
A refresh keeps the existing structure and technology. You're changing the surface, not the foundation.
A refresh is a good fit when the bones are sound. That means the site loads reasonably fast, you can edit the content you need to edit, the forms work, and Google can find your pages. What's wrong is cosmetic or strategic:
- The design looks five years old next to your competitors
- The homepage talks about the wrong thing, or buries what you actually sell
- Photography is weak or inconsistent
- The copy was written by committee and says nothing
- Mobile layout is cramped but functional
- You've added services and the navigation no longer matches the business
All of that is fixable without touching the underlying build. New type, new colors, better images, rewritten pages, reorganized menu, sharper calls to action. In our experience a focused refresh on a healthy site is where most of the easy wins live, because the biggest conversion problems are usually message problems, not technology problems.
The honest caveat: a refresh on a weak foundation is lipstick. If the site takes seven seconds to load, new fonts won't save it.
When a rebuild is the honest answer
A rebuild means new foundation. New code or new platform, new content structure, usually new hosting. You keep the brand and the good content and you throw away the plumbing.
Rebuild when the site is actively costing you time or money:
- Every small change requires a developer, so changes don't happen
- The site is slow in a way no plugin fixes, and speed is hurting search and conversions
- You're paying for a stack of plugins and subscriptions that only exist to patch gaps
- It's built on something unsupported or abandoned, or nobody knows how it works
- You need functionality the platform can't do: booking, quoting, gated client areas, a real product catalog, an integration with your CRM
- It breaks on phones, which is where most of your traffic is
- Security updates are behind and you've had scares
- You want to sell online properly and the current setup is a bolt-on
There's also the quiet signal we see most often: you've stopped using your own website. You don't send people to it. You don't update it. You work around it with PDFs, DMs, and spreadsheets. That's not a design problem. That's a site that no longer fits the business.
If your site started as a template and you've outgrown it, we wrote about that specific tipping point in when templates stop scaling.
The website redesign checklist
Run through this before you talk to anybody about building anything. Answer honestly, in writing. It takes an hour and it saves weeks.
Business questions
- What is the single most valuable action a visitor can take? Call, book, buy, request a quote, apply?
- How many of those happened in the last 90 days? If you don't know, that's your first fix.
- Who is the site mainly for? New prospects, existing customers, recruits, referral partners?
- What do people ask you on the phone that the site should have answered?
Content questions
- Which pages actually get traffic? Pull the top 20 from your analytics. Most sites have five pages that matter and forty that don't.
- Which pages do you want to keep, rewrite, merge, or delete?
- Do you have usable photography, or does that need budget?
- Who is going to write the new copy, and by when? This is the number one cause of stalled redesigns.
Technical questions
- What is it built on, and who controls the domain, hosting, and logins? Find out now, not during migration.
- How fast is it on a phone on cell data? Test it. If it feels slow to you, it's slow to customers. Core Web Vitals is the plain-English version of why that matters.
- What has to integrate? Payments, email marketing, calendar, CRM, inventory, accounting.
- What will you need to edit yourself, weekly? Be specific. "Everything" is not an answer.
Risk questions
- Do you rank for anything you can't afford to lose? If yes, URL mapping and redirects are non-negotiable in the scope.
- What's your real deadline, and what's driving it? A trade show and "sometime this spring" produce different projects.
If most of your answers are about content, message, and look, you need a refresh. If several answers are about things the site can't do, you need a rebuild.
A simple way to decide
Here's the rule we use when a business owner is on the fence.
Refresh if: the site does the right things badly. Rebuild if: the site can't do the right things at all.
Two more tiebreakers.
Add up what you spend per year keeping the current site alive. Hosting, plugin licenses, the developer you pay hourly for small edits, the time your team spends on workarounds. If that number is a meaningful chunk of a rebuild, the rebuild pays for itself faster than it looks.
Then ask what's changed in the business. New services, new pricing model, new market, a second location, a shift from one-off projects to retainers. Websites don't age out of fashion as fast as they age out of accuracy. A site built for the business you had in 2021 will quietly undersell the business you have now.
What each one costs and how long it takes
Rough shape, not a quote.
A refresh is typically a few weeks. The work is design, copy, images, and layout changes on an existing build. The variable cost is content: if you need new photography and all-new page copy, that's most of the budget and most of the calendar.
A rebuild starts around $2,500 for a straightforward marketing site and climbs with functionality. Custom booking, member areas, e-commerce, or integrations with the tools you already run on all add scope. We break down what drives the number in how much a custom website actually costs, and our starting ranges are on the pricing page.
One thing worth saying plainly: the budget item people skip is the one that matters most. Content. You can hire great developers and still launch a weak site if nobody wrote the pages. Decide early who owns the writing, and give them a deadline before design starts.
Don't throw away what's working
The most expensive redesign mistake is a clean slate. New sites often lose traffic for months, not because the design is worse, but because somebody deleted pages that were ranking and didn't redirect the old URLs.
Before anything launches, make sure your project includes:
- A full list of current URLs and where each one will point after launch
- 301 redirects for anything that moves or disappears
- Your analytics and conversion tracking carried over and verified
- Page titles and descriptions preserved on pages that already perform
- Forms tested end to end, including where the emails land
- A pre-launch check on real phones, not just a browser window resized
Also keep your winners. If one service page brings in half your inquiries, don't redesign it into something cleverer. Keep the structure, improve the clarity, and leave the thing that works alone.
Where to start
Pick one hour this week and fill out the checklist above, especially questions one, two, and five. The valuable action, how many times it happened, and your top 20 pages. Those three answers tell you more about whether you need a refresh or a rebuild than any opinion about how the site looks.
Then, if you want a second read on it, send us the list and the link to your site. We'll tell you straight which one we think you need, and if the answer is "refresh, and here's the one page to fix first," we'll say that too. Not every site needs to be rebuilt. The ones that do, you usually already know.