Whoever you ask whether your website needs a rebuild probably earns more from one of the two answers.

A web design agency makes its money on rebuilds. A freelancer selling monthly maintenance makes more when you keep patching the old site. Your nephew who "does websites" just wants something to put in his portfolio. And I have a stake too: Marshland builds websites, and a new build pays a lot better than a forty-minute fix.

So keep that in mind while you read this. Deciding when to rebuild a website, and when a redesign or a simple fix will do, shouldn't depend on who you asked. The questions below don't depend on trusting anyone, me included. You can answer every one of them yourself with a phone and five minutes.

Four questions that point to a fix, not a rebuild

Answer these honestly. Ask a friend to check the ones you aren't sure about. A "no" on any of them is usually a fixable problem, not a reason to start over.

  • Does it load? Open it on your phone over cellular, not your office Wi-Fi. If something shows up in under three seconds, that's a pass. If you're watching a blank screen, run this 10-second speed check before you decide anything. Most slow sites have one or two causes, and they're usually images or a bloated plugin.
  • Does it work on a phone? Can you tap the phone number and have it dial? Can you read the text without pinching to zoom? Does the menu open?
  • Can you find it in Google for your own business name? Search the exact name. If your site turns up on the first page, Google knows you exist. If Yelp, a directory, or a business with a similar name shows up and you don't, that's worth fixing, but it rarely takes a rebuild.
  • Does the form arrive? Fill out your own contact form right now and see whether the email reaches you. You'd be surprised how many sites have had a broken form for a year and nobody noticed. If yours didn't arrive, here's how to track down why.

If all four come back yes, your problem isn't the website. It's what's written on it. The site works, but the words don't explain what you do well enough to get someone to call.

Better writing fixes that. If you're not sure what the words should be doing, here's what a website should actually do for a service business.

Four that say redesign or rebuild

These are the situations where patching it costs more than starting over.

  • You can't edit it. Changing your hours means emailing someone and waiting a week. Or the editor is so tangled that you're afraid to touch it.
  • You can't get it back. The person who built it is gone and the login went to an old employee's email. That alone isn't a reason to rebuild: the renewal charge on your card tells you who the host is, and the registrar has a recovery process if you can prove the business owns it. Start with getting it back. It only belongs on this list if recovery fails or the site's files can't be retrieved.
  • It's built on something abandoned. A WordPress theme with no updates since 2019. A page builder that went out of business. A Flash site, which no browser has run since early 2021. Software nobody's maintaining stops getting security patches, and fixing it gets harder and more expensive every year.
  • It doesn't work on phones at all, and can't be made to. Some older sites were built with fixed widths that no stylesheet can rescue. Most local searches happen on a phone (your Google Analytics or Search Console device report will show your own split), so a site like this is sending most of your visitors away.

One of these on its own is a good reason to rebuild. Two of them settle it.

The middle case nobody talks about

Here's the one I run into most. The site is fine, but the business has changed.

Maybe you were a residential cleaning company in Mesa and now you do commercial contracts in Tempe and Chandler as well. Maybe you added a service, or dropped one, or hired two techs and want to take on bigger jobs. The site still describes the business you had three years ago.

This usually feels like a rebuild because the site feels wrong. But most of the time it's three new pages: one for the new service, one for the new area, and an updated homepage that says what you do now. That's a refresh, not a rebuild, and it's a few days of work.

What a rebuild actually costs beyond the invoice

The quote is the number everyone looks at. If you want a sense of fair ranges, here's what a small business website should cost. But the invoice leaves out a few things.

The first is your time on content. Every rebuild stalls at the same point: the designer is waiting on you for photos, service descriptions, and bios for the team page. Count on 10 to 20 hours of your own time even with a good writer helping. Most delays on a build come from here, not from the developer.

The second is the SEO risk window. When a site launches with new URLs, Google has to reconnect everything it knew about the old ones. Traffic often dips for a few weeks. If the build is sloppy it can drop for months.

The third is the redirect work most builds skip. Say your old site had a page at /services/ac-repair.html and the new one moves it to /ac-repair. Without a permanent (301) redirect, every Google result and old link pointing to that page now lands on a 404. Ask anyone quoting you a rebuild whether 301 redirects for every old page are included, and here are a few more questions worth asking. If they hesitate, that tells you a lot.

The same thing applies to the "free" rebuild your hosting company keeps offering. "Free" hosting has real costs, and usually they show up once you try to leave.

The one time to rebuild a site that works

There's one exception to everything above.

If you're afraid to touch your website, rebuild it.

I don't mean mildly annoyed. I mean you've had a holiday closure to post and decided not to bother because the last time you edited something the whole layout broke. Or you know your prices are out of date and you leave them, because opening that editor feels like defusing a bomb.

That fear costs real money. It's why the new service never gets a page, and why the Google listing and the website disagree about your hours. A site that passes all four "fix it" questions can still be failing you if you won't go near it.

In that case, what you're paying for in a rebuild is the ability to change your own site again.

Want a second opinion?

Before you send me anything, do one thing today. Take out your phone, turn off Wi-Fi, and fill out the contact form on your own website. Time how long the page takes to load, then check whether the email arrives within five minutes. If either one fails, you've found your first fix, and you haven't spent anything yet.

Then go through the eight questions above and send me your answers and your web address. I'll tell you which side you're on. If it's fix, I'll say so, even though a rebuild pays better. A lot of the sites I look at need fixes, not a rebuild, and I'd rather you trust me for the next ten years than overpay me once.