Start from the bias
A studio that builds websites makes more money from a rebuild than from a fix. Owners know this, which is why "you need a new website" is heard with suspicion, and rightly. So the method for deciding has to be one you can check, with the reasoning written down, and it has to be able to conclude "fix" most of the time, because that is the honest answer most of the time.
Here is ours.
Step 1: the four disqualifiers
Before any cost comparison, four findings make rebuild the answer on their own. If none applies, we are in fix territory whatever the site looks like.
The platform is dead. A website builder that has shut down or stopped development, a theme whose author disappeared years ago and which will not survive the next platform update, a framework several major versions behind with no supported upgrade path. You cannot fix what cannot be updated.
It was never built for phones. Not "the mobile layout has a fault" but "the site was designed for a desktop screen and mobile was never considered". Retrofitting a responsive layout onto a fixed one costs more than starting again.
Security cannot be trusted. A site that has been compromised in a way where cleaning it would leave doubt, or one running software so old it cannot be patched. You do not rebuild trust in a codebase; you replace it.
The structure is wrong for the business. A brochure site trying to become a booking business, with no concept of services, staff or prices in its structure. Bolting booking onto that produces something worse than either. When the business has changed shape, the site's skeleton has to as well.
Any one of these, and the diagnosis says rebuild, with the reason. Absent all four, we go to step two.
Step 2: the list and the numbers
We list every fault found, with a fixed price for each. The typical list on an inherited site: dead form, oversized images, plugin cleanup, database cleanup, a mobile layout fault, email authentication, a missing padlock, no admin access. Each has a price and most are small.
Then two totals:
Fix total: the sum of the fixes, plus the first year of a care plan so the faults do not return.
Rebuild total: the package price for a new site, plus the first year of a care plan, plus the one cost owners forget: the risk to search rankings when URLs and content change, which is real and which takes months to recover from even when done carefully.
For most sites, the fix total is a third to a half of the rebuild total, and it has no ranking risk. That is why the answer is usually fix.
Step 3: the two-year view
The fix total can be misleading if the site is going to need fixing again next year. So we ask what the site will cost over two years under each route.
A fixed site on a care plan should need nothing beyond the plan. A fixed site with no care plan will decay again, and the two-year cost includes the next round of fixes. A rebuilt site on a care plan is the most expensive route by far in year one and the cheapest thereafter, but "thereafter" is a long time to wait for a return.
The honest picture for most businesses: fix now, care plan from now, and revisit rebuild in two or three years when the business has changed enough to justify it, if it ever does.
Step 4: what the owner wants the site to do
One question changes the arithmetic more than any other: has the business changed? A salon that added three stylists and now wants online booking with deposits has different needs from the salon the site was built for. A clinic that opened a second location needs location pages the old structure cannot hold.
If the business has changed and the site's structure cannot follow, that is the fourth disqualifier arriving through the side door. If the business is the same and the site is simply neglected, fix it.
What we write down
The diagnosis report has one page that says, in order: the four disqualifiers and whether any applies; the list of faults with a price each; the fix total and the rebuild total over two years; and a recommendation with the reason. You can take it to another studio, and we tell clients to if they want a second opinion, because a recommendation that survives a second opinion is worth more.
Most of them say fix. The ones that say rebuild say why in a sentence you could repeat to your accountant.
Where this sits
This is what the $299 diagnosis produces. It is priced so that the decision costs less than getting it wrong in either direction, and it is yours regardless of who does the work afterwards.