You rebuild a website when keeping it costs more than replacing it — not when it turns five, and not because you have stopped liking it. Six signals justify a full redesign: it is unusable on a phone, it cannot be edited without a developer, it runs on unmaintained technology, it has stopped converting while traffic holds, it cannot carry your current offer, or somebody else holds the keys. Any one of the six is enough.
Three other reasons come up constantly and justify nothing at all. This piece sorts them out, gives the rule for choosing between repair and rebuild, and sets out real Quebec costs.
The six signals that justify a redesign
1. It is unusable on a phone. Not “it looks a bit off”: it forces pinch-zooming, the menu will not open, the form cannot be completed with a thumb. Most of your visitors arrive on mobile; a site that loses them loses most of its traffic before the first line is read.
2. You can no longer change anything yourself. If updating a phone number takes a quote and three days, the site is not a tool any more, it is a dependency. This one is easy to measure: count the changes you wanted to make this year and did not.
3. The technology is no longer maintained. Abandoned plugins, an outdated PHP version, a theme whose author has vanished. This is not an aesthetic question but a security one: an unmaintained site eventually gets compromised, and the cleanup costs more than the rebuild.
4. Traffic holds but enquiries collapse. This is the most reliable signal, because it is a number. If visitors have been flat for two years while form submissions halved, the site no longer meets its visitors’ expectations. One caution: before concluding, check it is not a single localised leak — the method is in conversion funnel leaks.
5. The site cannot carry your current offer. You have added three services, opened a second territory, changed who you sell to, and the 2019 structure has no room for it. Stacking pages onto an unsuitable architecture produces a site nobody understands — neither visitors nor Google.
6. You do not own your access. Domain registered in the agency’s name, hosting you cannot reach, no exportable backup. This is not a technical problem, it is an asset problem: as long as it lasts, your online presence belongs to somebody else.
Rebuild when keeping the site costs more than replacing it.
The three false signals
“It is five years old.” Age is not a criterion. A well-structured, fast, up-to-date site from 2020 outperforms a rushed 2026 rebuild. Calendar-driven redesigns are an agency habit, not a business need.
“I am tired of looking at it.” Understandable — you see it every day, your visitors see it once. Owner fatigue is the most common engine of unnecessary redesigns and the least often admitted. If the only complaint is aesthetic, refreshing colours, type and photography costs about 10% of a rebuild.
“My competitor’s is nicer.” A nicer site does not sell more. Before copying, ask whether theirs converts: you will never know. What you can compare is how clear the offer is in eight seconds — and that can be fixed without rebuilding anything, along the structure described in the home page that converts.
Repair, restructure or rebuild
There is a middle level almost nobody offers, and it is often the right one. Repair: keep everything and fix what blocks — speed, forms, home page copy. A few days, a few hundred dollars. Restructure: keep the technology and the content, redo the architecture and the design. Two to four weeks. Rebuild: everything new, hosting and copy included.
The sorting rule: if the signals are mostly about content and presentation, restructure. If they are about technology, security or who owns the access, rebuild. You do not restructure a site built on an abandoned plugin.
What it costs, and what it will not fix
Our pricing is public: $800 for a one-page site, $2,800 for a brochure site, $8,500 for a full site, $12,000 for a transactional one, plus maintenance from $120 a month. Market ranges across all suppliers in Quebec are in what a website costs in Quebec, and the construction-sector case in the contractor website.
What a redesign will not fix: a lack of traffic. A brand-new site on 120 visitors a month is still a 120-visitor site. A redesign improves the pass-through rate, not the volume coming in — that is stage 2 of the sales funnel, never stage 1.
And above all: a badly prepared rebuild loses rankings that took years to earn. That is the main risk, it is entirely avoidable, and the procedure is in migrating a site without losing your SEO. If you are torn between repairing and rebuilding, book the call: we run the six signals together and the answer lands within fifteen minutes.