Fix the site if it is editable, works on a phone, and is wrong in ways you can point at. Start over if you cannot edit it, it was built before phones mattered, or fixing it needs more than about three separate repairs. That is the whole decision; the rest of this article is how to answer those three honestly, and what a rebuild must keep from the old site so you do not lose what was working.
Seven questions, in order
Answer each with yes or no. Stop when the answer points clearly one way.
Question 1 — Can you edit it? Can you, or someone you pay by the hour, change a sentence, a price or a photo today? If the site is on a platform nobody remembers the login for, was built by someone who has moved on, or needs a developer for a text change, it is not editable in any useful sense. A site you cannot edit is a site you cannot fix. Start over.
Question 2 — Does it work on a phone? Open it on your phone. Does the text fit the screen without pinching? Are the buttons tappable? Does the menu open? If the site shrinks the desktop page down to fit, it predates mobile and no amount of patching changes its bones. Start over.
Question 3 — Is it slow in a way you can feel? Over mobile data, off Wi-Fi: how long until you can read the headline and tap the phone number? A slow site with a modern platform is usually fixable (photos too large, a plugin too many). A slow site on an old theme usually is not, because the slowness is the theme. If it is the platform, start over; if it is the photos, fix.
Question 4 — Does it say what you do and where? If the headline is a slogan and the service area is missing, that is a writing problem, and writing problems are the cheapest fix there is. Fix.
Question 5 — Can people reach you from it? A phone number that is not tappable, a form that delivers nowhere, a contact page three clicks deep. All fixable on an editable site in an afternoon. Fix.
Question 6 — Does it look like your business? Your own photos, your real reviews, your name. If the site has none of these, adding them is editing, not rebuilding. Fix. If it cannot show them because the layout has no place for them, that is a structure problem; see question 7.
Question 7 — How many of the above failed? One or two fixable items: fix them. Three or more, or any of questions 1 to 3: the repairs will cost more than a site that starts right, and you will still have the old site's limits afterwards. Start over.
What a rebuild must keep
A new site is not a blank slate, and treating it as one is how businesses lose the Google standing their old site had earned.
- The addresses that rank. Any page that brings visitors from search keeps its address, or the old address redirects to the new page. Ask your builder to list the pages they will redirect before they start.
- The reviews and the photos. These are the most convincing things on the old site. Move them, with their sources.
- The phone number and the business name, exactly as they appear on your Google listing. Mismatches confuse Google and customers alike.
- The pages that were working. If one service page was producing calls, the new version is a cleaner copy of it, not a reinvention.
What a rebuild must change
Everything that failed questions 1 to 6: an editable platform, a mobile-first layout, honest speed, a headline that says the trade and the town, contact in the first screen, and places for your own photos and reviews. What a local business homepage needs is that list in the order a visitor meets it. If you are not sure which of the six reasons in why your website isn't generating leads applies, the rebuild should fix all of them at once, which is the main argument for doing it.
Do not rebuild to fix a traffic problem
One honest caveat. If nobody is visiting the old site, a new site will not be visited either. Being found is a Google Business Profile and page-title job before it is a design job. Fix the listing, fix the titles, then judge whether the site converts the visitors it gets. Website traffic but no leads is the case where the visitors are there and the site is the problem.
Let someone else answer the seven questions
The free website audit runs the machine-checkable half (mobile-friendly, readable by Google, title and headline present, a way to call you, broken links) and the report shows which of them fail. If it is a rebuild, the pricing page says what that costs and what is included. If it is three fixes, you will be told that, because telling you otherwise would be a short-lived way to win a client.
For a trade-specific view of what the rebuilt site should contain, see websites for roofers or websites for dentists.