Fix it if the structure underneath is sound and the problems sit on the surface. Rebuild it if you cannot change the things that need changing, or if every fix has to route around a decision somebody made three years ago.
That sounds obvious until you are the one holding a quote. A rebuild is the option most builders reach for, because it is easier to price and more interesting to build. Sometimes it is the right call. Often an owner is sold a new site when what they needed was two days of work.
How do you tell which one you need?
Stop looking at whether the site looks dated and write down what is actually failing.
Not "it feels old". Specific things. The phone number is wrong on three pages and I cannot fix it. The contact form does not reach me. It takes forever to load on a phone. I cannot add a fourth service without paying someone. Nobody finds us when they search our town.
Then sort that list into two piles: things that are wrong on the page, and things that are wrong underneath it.
Wrong on the page is content. Copy, photos, prices, a missing button, a page that was never written. Wrong underneath is structure. What pages exist and how they relate, what you are allowed to edit, how fast the site can possibly be, whether it works on a phone at all, whether search engines can read it.
Surface problems are fixes. Structural problems are rebuilds. The confusing part is that the two are unrelated to appearance. A site can look rough and be structurally sound. It can also look perfectly pleasant and be unfixable.
What is genuinely worth fixing?
More than most quotes admit.
Bad copy is a fix. If the homepage does not say what you sell or where you work, that is a writing job, not a build job. The same site with the right words on it converts differently, and that is the cheapest change available to you.
Missing pages are usually a fix. If you have five services and one page covering all of them, adding four pages is work on top of what you already have rather than a reason to start over.
Broken plumbing is a fix. Forms that do not deliver, a wrong address in the footer, a booking link that points somewhere dead. None of that needs a new site. It needs someone to open it and repair it.
Photos are a fix. Stock images of people who do not work for you are replaceable in an afternoon, once you actually have pictures of your own work.
Local visibility is mostly a fix. Consistent name, address and hours, a service-area page per town, a Google Business Profile that matches the site. That work sits on top of almost any platform.
What actually means rebuild?
Four situations, and they are all structural.
You cannot edit what you need to edit. If changing a price is a developer ticket, and it will always be a developer ticket, no amount of patching changes that. The cost is not the one change, it is the year of changes you will not make.
The platform is gone or going. The builder shut down, the theme has not been updated in years, the plugin the site depends on has no maintainer. You are not fixing a site, you are keeping something on life support. That is the same trap described in leaving a website builder.
It does not work on a phone. Not "it looks a bit cramped" but genuinely does not work: text you have to pinch to read, buttons you cannot hit, a menu that does not open. Retrofitting that onto a desktop-only build usually costs more than building it right the second time.
The business changed and the site did not. You sell something different now, or to different people, or in a second language. A site organised around what you did in 2022 will fight you on every page. Structure follows what you sell, and if the two have diverged, editing text will not close the gap.
If none of those four describe you, be sceptical of a rebuild quote.
Is a slow site a rebuild?
Sometimes, and it is worth finding out which before you spend anything.
Slow has cheap causes and expensive ones. Enormous uncompressed images, a stack of tracking scripts, a page loading fonts nobody uses: those are fixes, and they are often the bulk of the problem. Slow hosting is a fix too, and a cheaper one than people assume.
The expensive cause is a build that has to do too much work before anything appears on screen. A page builder stacking effects, a theme carrying features you do not use, a site assembling itself in the browser on every visit. You can trim that. You cannot usually make it fast.
The order matters: measure, remove the cheap causes, measure again. If the site is still slow after that, the slowness is structural and it belongs on the rebuild side. The full version of that diagnosis is in why your website is slow.
What does each option cost?
Different enough that guessing wrong is expensive in both directions.
Fixes are hours. Copy rewrites, new pages on an existing structure, form repairs, image work, local search cleanup. You can buy them one at a time and stop when the problem is solved.
Rebuilds are projects. Our marketing sites and landing pages run from $1,200 to $8,000, and where a project lands depends on how many pages there are, how much of it you need to edit yourself, and whether it ships in one language or two. Store and B2B builds start at $2,900. Those numbers live on pricing, and how we scope them is on web design. What drives that range up and down is broken out in what a website costs in Ontario.
The mistake that costs the most is spending rebuild money on a site whose real problem was that nobody had written the copy. The second most expensive mistake is spending two years of small fixes on a site that was never going to work, which quietly comes to more than the rebuild would have.
What happens to the rankings you already have?
This is the question that should slow a rebuild down, not stop it.
A site with real search traffic has value that is not visible in how it looks. Rebuilding without carrying that over is how a business loses enquiries for months and cannot work out why. Every URL that changes needs a redirect, page titles and headings need to survive the move, and the content that was actually ranking needs to still exist on the new site.
None of that is a reason to keep a site you cannot use. It is a reason to hire someone who treats it as part of the job rather than an afterthought, and the checklist for it is in redesigning without losing rankings.
If the site has no traffic to protect, this whole worry disappears and the decision gets simpler.
Is there a middle option?
Yes, and almost nobody offers it, because it is harder to sell than a rebuild.
Keep the structure, replace the surface. New copy, new photos, the pages you are missing, the plumbing repaired, speed cleaned up where it is cheap to clean. You find out in six weeks whether the site was the problem, and you find out for a fraction of the money.
If enquiries improve, you were right to fix it. If nothing moves after the surface is genuinely good, you now know the problem is structural and you are buying the rebuild with evidence instead of a hunch. Either way you need to be able to see the difference, which means knowing what to measure before you start: how to tell if your website is working.
The version of this to avoid is the drift. A little work every few months, no measurement, no decision, and three years later you have paid for a rebuild in instalments and still have the old site.
If you cannot tell which side of the line your site falls on, send us the URL for a free teardown. We will tell you what it is built on, which of your problems are surface and which are structural, and whether we would fix it or rebuild it, whether or not you work with us.