Two different projects hiding inside one word
When an owner says the website needs work, one of two complaints is usually underneath. The first is cosmetic: the site looks dated, the photos are old, the logo changed two years ago and the site never did, and it is faintly embarrassing to send a customer there. The second is structural: the phone is not ringing from search, the site has one page trying to cover six services and eleven towns, and no amount of new paint changes that math. A redesign answers the first complaint — new design layer, same underlying pages, same addresses, same structure. A rebuild answers the second — new architecture: a page for every service, a page for every market that matters, new internal links, often new page addresses.
Confusing the two is where the money leaks. Owners regularly pay rebuild prices for what turns out to be a template swap on the same thin structure, then wonder why the leads did not change — nothing about the site's coverage changed, so nothing about its rankings could. Just as often, an owner with a genuinely broken structure buys a cheap redesign, because it was the smaller quote, and spends another two years not ranking with a prettier site. The packages question is never really 'how much for a website.' It is 'which problem do I actually have.' These five tests answer that before you talk to anyone selling either project.
Test 1: Count the pages against the searches
Write down every service you sell and every town you serve profitably. Multiply. That product is roughly the number of distinct searches your customers run — roof replacement in one town is a different search, with different results, from roof repair in the next town over. Now count how many pages on your site are dedicated to exactly one of those combinations, or at least to one service or one market specifically. If the site has a page per service and real coverage of your core towns, its structure is sound, whatever it looks like. If one services page and a bulleted list of towns is carrying the whole business, the structure is the problem, and structure is rebuild territory.
This is the test that decides most cases by itself, because Google ranks pages, not companies. A redesign cannot add the pages the searches require; it can only restyle the shortage. Our breakdown of how many pages a service company website should have walks the counting in detail, but the short version is brutal: if the pages do not exist, the rankings cannot, and no color palette fixes an absence.
Test 2: Are the current page addresses worth anything?
Open Google Search Console — or ask whoever manages your site to — and look at which pages earned impressions and clicks over the last few months. If your service pages show up there, those addresses hold real, accumulated value: history, links, and a track record in Google's index. That value is an argument for a redesign that keeps the addresses exactly as they are, or for a rebuild that treats address changes like surgery, with a redirect from every old address to its closest new equivalent. If, on the other hand, Search Console shows almost nothing — a handful of impressions on the homepage and silence everywhere else — the addresses hold little, and a rebuild risks almost nothing.
The disaster case is the accidental rebuild: a redesign project that quietly changes every page address because the new template or platform structured things differently, with no redirects, because nobody scoped them. The owner bought paint and got demolition. It is the single most common cause behind the pattern in our checklist on why Google rankings drop — and it is entirely a scoping failure, not a design failure. Whichever project you buy, the sentence 'page addresses stay the same, or every changed address gets a redirect' belongs in writing before work starts.
If part of your revenue comes from commercial accounts, our guide to website development for commercial service companies explains what property managers and procurement teams look for before they request a proposal.
Test 3: Can the current platform even hold the fix?
Some sites cannot be redesigned into health because the platform underneath forbids it. The warning signs: you cannot add pages beyond a template's fixed set, you cannot control page titles and headings properly, the site is locked inside a rented website program where the pages live on someone else's system, or the 'website' is a single scrolling page the builder made in an afternoon. If the platform cannot hold a page per service and per market, custom metadata, and a real contact path on every page, then redesign money spent on it is money spent decorating a ceiling you will hit within the year.
The honest version of this test cuts the other way too. Plenty of service companies run on ordinary platforms that can hold everything they need — the site is just old, slow, and thin in a few places. That is a redesign and a content expansion, not a rebuild, and any seller who tells you otherwise without walking your page structure first is selling the bigger ticket, not the right one.
Test 4: Does it work on a phone, under pressure, in thirty seconds?
Most service searches happen on a phone, many of them in a moment of mild emergency — water somewhere it should not be, a tree leaning wrong, a house that will not cool. Time yourself, or better, hand your phone to someone who has never seen the site: can a stranger confirm you do the service they need, in their town, and reach a short form or chat in under thirty seconds? If the bones do that and only the look fails, redesign. If the path itself is missing — no form, no service confirmation, a phone number as the only door — the problem is conversion structure, and conversion structure is built, not painted.
This is also where the two projects legitimately combine. A rebuild that keeps a confusing contact path has wasted its budget, and a redesign that sharpens the contact path on a sound structure can pay for itself without touching architecture. Scope the project to the failure. Sellers scope to their product line; owners should scope to the diagnosis.
Test 5: Who writes it — and who owns it afterward?
Ask both kinds of seller the same two questions. First: who writes the words? A redesign quote often assumes your existing text slides into the new design, which means your thinnest asset — one paragraph per service, written years ago — survives the project untouched. A real rebuild includes writing: pages that explain each service, each market, and the questions customers actually ask. Second: when it is done, what do you own? The domain, the site, the content, the ability to leave. Projects priced suspiciously low sometimes stay cheap because the site is rented, and the rent ends with the site disappearing. Neither question is about design taste, and a seller's answers tell you more than their portfolio does.
Compare the quotes on those terms — page list, writing included or not, addresses preserved or redirected, ownership in writing — and the price gap usually explains itself. The cheaper bid is frequently the smaller project wearing the same name. Our guide to choosing between a builder, an agency, and a done-for-you build covers the seller side of that comparison; this test is the project side.
The honest answer, and where ProRouteX lands
If your site has a page per service, real town coverage, addresses earning impressions, and a contact path that works on a phone — redesign it, keep the addresses, and spend the savings on reviews and coverage gaps. Anyone who sells you a rebuild on top of that structure is selling, not advising. If the pages do not exist, the platform cannot hold them, or the site earns nothing worth preserving, a redesign is the expensive option no matter how small the quote: it spends real money to leave the actual problem standing.
ProRouteX is a rebuild shop, and we will say plainly that not every site needs us. What we build is the structure itself: service pages, city pages, profile and schema work, one company per trade, per market. The Starter package starts at $997 — up to 10 service pages and your top 3 city pages, delivered in one week — with wider systems quoted on the market, the optional Growth Engine plan at $497 a month, and you own the site and content outright. Run the five tests on your current site first. If it passes them, keep it and redesign. If it fails the first one, you already know which project you are buying — the only question left is who builds it.
