Support · Websites · 17 min read
How long a Canadian website actually takes
A focused marketing site is usually six to ten weeks: about eight pages, a CMS your team can edit, and a calendar you can run even if you never hire us.

The decision this week: this quarter, or a later phase
If you need a Canadian marketing site this quarter, the useful decision is not “make it modern.” It is this: can you ship about eight honest pages in six to ten weeks, in CAD, with one conversion a stranger can finish? A custom website launch at this studio starts at about $8,500 CAD for that shape — information architecture, a CMS operators can edit, bilingual routes if you buy them, analytics, twelve months of hosting, and a 90-day defect window after the domain cut. You can use the same fence if you never hire SYCGeeks: eight pages, one conversion, one signer, copy in hand. Everything else is a later phase you can date.
Write the conversion as a sentence a receptionist could repeat: “A nurse in Surrey books a Thursday cut on Square.” “A buyer in Toronto requests a CAD quote for a bilingual packaging run.” “A clinic in Burnaby sends an intake form that lands in the same inbox every time.” If two conversions fight — a newsletter and a booking, a store and a lead form — pick one for launch and put the other on a dated line. SYCGeeks has priced this way from Vancouver since 2015. The cheaper path is a site a stranger can finish, not eight calls to action that split the first visit.
Hard dates belong in the first note: a market in May, a fiscal year, a lease, a college advertising rule, a federal bid. If you need something live in three weeks and you have no copy, you can still ship a thinner first version — home, contact, one service page, legal shell — and schedule the rest. That is cheaper than promising eight finished pages and sliding in silence. Write admin@sycgeeks.ca with the conversion, the date, the languages, and Square or Shopify if you already pay for them. Prefer to talk? Book a Square appointment on Pacific time; no studio account is required. Toronto, Montreal, and Halifax clients are welcome; the overlap with Pacific hours is the only constraint.
Builds at this baseline bill in four milestones: 30 percent to start, 30 percent when architecture is approved, 30 percent when the core works, 10 percent at launch. That split is useful even on a project you run internally. Pay (or allocate staff time) against named gates, not against a vibe. If the brief grows — a second language you did not plan, a checkout, a members area — write a new number and a new gate. Folding extra weeks into the original CAD figure is how teams lose track of what “done” means. You should always be able to say whether you are inside the signed eight pages or looking at a dated add-on.
What six to ten weeks and about eight pages actually buy
Count the pages on a napkin: home; two or three services; how you work or about; a proof or guides index; contact or booking; privacy; terms. That is eight. A custom website launch here is “up to 8 pages on 4 unique templates.” The templates matter more than the page count — home, a service layout, a long-read, a legal/contact shell — because operators will clone them later. If you already know you need twelve service lines, pick the three that pay for the quarter and park the rest. Extra pages after launch are typically one to two weeks each. Starting at eight is the cheaper path; inventing twenty URLs in week one is how six weeks becomes fourteen.
The first two weeks are structure, not decoration. Write the sitemap, the one conversion, and the pages that must exist on day one. Decide English only, French only, or both — that sentence changes button widths, legal copy, and the calendar. If both languages must be true on day one, budget a language pass on the critical path, not a plugin weekend. If you only have English, you can still reserve French routes (/fr/…) and ship a complete first language. Publishing a machine dump so the homepage can show a flag costs more later than waiting for a signer.
Weeks three through six are design and build on the stack your operators can keep. For teams that want a design system they can keep, that is Next.js. Choose after you meet the person who will change a phone number in February. A stack that impresses a developer and terrifies the office manager is a failed project at month four — you will call +1 (672) 200-7924 for work a CMS should have allowed. Core Web Vitals belong in this stretch: one H1, readable URLs, retina crops, templates that stay fast after someone adds a third paragraph.
The last stretch is content, bilingual review if French is in scope, legal pages, redirects, and launch. That stretch is where honest calendars slip, and it is rarely because of a missing component. A privacy page that still mentions CCPA and a California address, a fee list living in a spreadsheet, or a partner who wants “one more round” will add a week faster than code. PIPEDA and WCAG sit here too: form labels, contrast, a keyboard path, a privacy sentence that says what you collect and why, in the language of the page. Chat and analytics are still personal information. Health, legal, and financial teams should send college or insurer rules in week one so the templates can hold them.
A week-by-week path you can hand to a partner
Week one is intake: signed CAD scope (or an internal one-pager with the same fence), access to the current domain and analytics, and a 90-minute working session. Leave with a one-page sitemap and the conversion written as a sentence. Name the signer and a backup. Week two is architecture: URLs, the CMS model, launch pages versus later pages. If French is in scope, the route plan is this week even if French copy arrives in week five. Waiting to “add language later” is how you ship a folder tree that cannot hold Quebec. You can run both weeks as a shared document. A studio dashboard (intake, active, review, paused, launched) is just that document with statuses.
Weeks three and four are the first templates: home, one service page, contact or booking, and the legal shell. Approve in a short written note, not a 40-comment PDF. Weeks five and six fill the remaining pages and the CMS training. The operator should change a heading and a phone number in front of someone before you call the build done. If they cannot, you picked the wrong stack or the handoff is incomplete — cheaper to fix now than in February. Firms that arrive with copy in week one often launch near week seven. Firms that discover their fee list is a rumour in week six launch near week ten, or they pause. Neither is a failure if you wrote why.
Weeks seven to ten are the honest buffer: remaining content, bilingual review, redirects from the old site, a real enquiry test, DNS, SSL, GST/HST behaviour if anything is sold. If the signed scope is a store or an API, keep this picture for the public site and add a second picture for money. Ecommerce is quoted at eight to fourteen weeks (from $14,500 CAD). A custom API is six to twelve weeks (from $18,000 CAD, up to three named integrations). A first mobile app is twelve to twenty weeks (from $28,000 CAD, about 8–12 screens). None of those hide inside a marketing-site quote. If someone told you the shop “just plugs in,” they were describing a theme demo, not a Canadian checkout with provincial tax.
Checkout, tax, inventory, and webhooks have their own approvals — often finance and a Square or Stripe owner who is not the marketing partner. Those people need calendar invites in week one of that phase, not a forwarded message the Friday before you want to take a card. Sequence on the same project record so you keep the history: public site this quarter, store next quarter. Care members pay about 25 percent less on published add-ons when the second phase is a scoped change to a live product. Monthly care is $450 CAD; yearly is $4,500 CAD (two months included versus paying month by month). You can skip care if someone on staff will watch hosting and the form; the cheaper path is naming that person, not hoping Monday will notice a weekend outage.
Content and one signer move the date — here is how to tell
The delay you will actually feel is waiting for words and pictures that make a Canadian page true. Service descriptions still in a partner’s head. Headshots from 2019. A fee schedule finance has not signed. Legal pages copied from an American template. You can build the templates either way. You cannot invent prices or professional constraints. A useful test in week one: can the signer paste the fee list, three service paragraphs, and a privacy purpose into a shared folder? If the answer is “we will circle back,” the short end of six to ten weeks is already gone. Treat it as a calendar fact. Put the date on the worksheet and choose pause or a thinner launch.
Assign one approver who can change a headline without a second meeting. Put their name in the brief. Name the backup before you start, not the Friday before launch. A committee of five that all need to weigh in will add about two weeks to a six-week site; you can tell because every file waits on the next monthly partners’ meeting. The cheaper path is a standing 20-minute Friday note: what shipped, what is blocked, whose name is next to the missing PDF. You can run that in email. If you work with a studio, ask for the same note in the project record so a silent week cannot hide.
Send what you already have in the first week: the current site, the PDF you hand people, the Square item names, the Google Business hours, the French phrases you already use on voicemail. Screenshots of the page that fails beat adjectives. “Make it modern” is not a brief. “This booking form loses people after the phone field” is a brief. Price and schedule against the second kind, in-house or not. If you are rewriting an old site with years of untagged posts, say whether you also need ten years of untagged posts. Rebuilding on a clean set of blocks can still fit six to ten weeks; migrating a museum usually cannot. Redirect the URLs that still earn a real enquiry and leave the rest.
If content is late, keep the code calendar honest: pause, write the blocker, restart when the file arrives. Launching on placeholder Latin and stock handshakes is the expensive path — you will live with a draft, then pay again to replace it. Holding the domain cut until the privacy page names the tools you actually use is cheaper, even if the flyer already went to print with a “coming soon” URL. After launch, the 90-day window is for defects (a form that drops French submissions, a redirect that 404s the old blog). A new members area you thought of at a party is a quote on the same project: extra page one to two weeks, language pass one to two, checkout change two to three.
How to tell a store or an API is a second phase
Money and other people’s systems add weeks because they need written tax behaviour, a real CAD test, and an owner who is not marketing. Square, Stripe, or a warehouse feed should not be rushed into a Friday deploy because a campaign starts Monday. GST/HST by province, shipping to some provinces and not others, a product taxable in Ontario and exempt in another case — those are decisions, not theme settings. Test a real CAD order and a refund before launch. A demo card in US dollars is not a test. You can tell you are still in marketing-site scope if the site only collects leads and never takes a card. You can tell you have crossed if a “Buy” button appears, even for a deposit or a gift card.
If you need a public site this quarter and a store next quarter, say that in the brief and keep one project history. Care on the live site makes the second phase cheaper to attach. If partners, a mobile app, or a warehouse must read and write your catalog, that is not “a small feed” in week eight — it is a product with versioned routes and webhooks, starting at three named integrations. The cheaper path is connecting to the system you already pay for (Square inventory, Shopify, a warehouse file) rather than inventing a second system of record. Double-sold Saturdays almost always come from two books that disagree.
Bilingual review sits on the critical path the same way. French is not thinner English. Buttons that fit “Book” may not fit “Prendre rendez-vous.” Legal sentences change length. Design the language switch so people keep their place, and test forms in both languages before launch. A checkout that fails in French is a product bug, not a translation ticket for later. Name the reviewer who can sign French; it is often not the English signer. A bilingual associate in Montreal and a managing partner in Vancouver will disagree about tone. That disagreement is week-two work. If nobody can sign French, ship English complete and keep the routes ready — cheaper than inventing a voice for a regulated profession.
Hosting, the Canadian domain, and SSL belong in launch, not a surprise invoice. A staging URL and a handshake is not a launch. GST/HST on a studio invoice is not the same problem as GST/HST on your checkout, but both need a grown-up before you cut DNS. If the site sells anything, decide the provinces and the tax behaviour in writing. If it only collects leads, say that so nobody leaves a fake “Buy” on a service page. Canadians notice when a price is missing or when a dollar sign looks American. Pacific time is the clock on our appointments; say so on the site if clients are in other provinces.
Launch week, 90 days after, and how to start
Launch is a status change you can checklist: cut DNS, confirm SSL, place a test enquiry, watch the first real one land in the inbox or the Square calendar. If you take bookings, the appointment must appear in the dashboard, not only in email — email-only confirmation is how Tuesday double-books. Have someone at the business who can answer admin@sycgeeks.ca and the studio line on launch day. We work weekdays, Pacific time. If production is down — checkout, the form, the domain — call +1 (672) 200-7924. That line is for heat. A typo in a footer can wait for the notes field. Mixing those two is how urgent work gets lost in a thread about kerning. You can use the same split internally: phone for outage, a note for copy.
After the cut, stay on a 90-day defect window and write what changed. If you want a second page, a language you did not buy, or a checkout change, that is an add-on on the same launched project. Reopen the record; you should not have to explain the business again to a new account manager. A launched site without someone watching hosting and the form is a weekend problem waiting for Monday. Care is one way to buy that watch: hosting, small copy and dependency updates, one reserved studio hour, a dashboard note. The reserved hour is for small honest work. Eight reserved hours exist for work bigger than the monthly hour and smaller than a new build. If you are unsure which bucket you are in, ask — the answer should be a sentence.
A useful first sentence to start is specific: “We are a two-location clinic in Vancouver and Burnaby, we need English and French service pages, we book on Square, and the college site must mention us by November.” That is enough to price a six-to-ten-week site or to tell you the store is a second phase. A messy brief is still usable if it names conversion, deadline, languages, and systems. An empty brief is not — there is nothing to schedule. The calendar starts when the facts are in the room, not when the logo files arrive. If production on an existing site is down today, call. If you would rather talk than write, book. Those are the three doors.
If you already have a Next.js app that only the last contractor understood, send the repository in the brief. Inheriting a real system is cheaper than inventing a second one beside it. If you already have a site that is a pile of abandoned builders, a clean Next.js rebuild can still fit the window if you are not also migrating every post. Pick the stack for the people who will edit it, measure the real service page on a phone, and keep CAD on the invoice and on the page. That is the whole calendar, whether we build it from Vancouver or you run the worksheet with your own team.
Timeline worksheet: fill this before you sign a week-one date
Copy this list into a note and put a name and a date next to each line. Conversion in one sentence. Hard date (or “no hard date — target week ten”). Languages on day one: EN, FR, or both. Signer and backup, with leave dates. Current domain, Google Business URL, analytics access. Systems you already pay for: Square, Stripe, Shopify, a warehouse tool. Page list at eight or fewer, each with an owner for copy and an owner for photos. Legal: privacy purpose, terms, refund if you take money. Tax: leads only, or provinces you will charge GST/HST. One primary action per page. French signer if FR is in scope. College or insurer PDF if you are regulated. This list is the cheaper path because it surfaces the real calendar before anyone opens a repository.
Files you can attach this week, even if they are ugly: the current site export or screenshots; the PDF you already hand to people; a spreadsheet of fees in CAD; three photos that are actually yours; the voicemail script in both languages if you have one; the last invoice so currency and legal name match the site. Missing files are fine if they have a date. Invisible files are not — they are how week seven becomes a surprise. If half the cells are empty, you can still plan: those empty cells are the buffer. If the whole list is empty, you are not in a six-to-ten-week project yet. You are in a discovery week, which you can buy as a scoped conversation or run yourself with this worksheet.
Scope fence to paste at the top of the note: “Launch is eight pages, one conversion, CAD invoices, GST as applicable, Pacific-time support weekdays. Store, membership, custom API, and extra languages beyond what we named are dated later. Architecture approval is milestone two (30 percent). Core templates working is milestone three (30 percent). Domain cut is milestone four (10 percent). Pause is allowed when named files are late; placeholder launch is the expensive option.” You can change the numbers; keep the gates. Partners understand gates. They do not understand “it depends” in week nine.
How to use the worksheet this week without waiting on a studio: fill the list on Monday, book 45 minutes with the signer on Wednesday, send the missing-file dates on Friday. If a file cannot have a date, drop that page from launch. If French cannot have a signer, drop French from day one and keep the routes. If money cannot have a tax owner, drop checkout from this phase. That is resourcefulness, not shrinkage. The site that ships on week eight with a booking form that works will outperform the site that almost included a store. When you are ready for a CAD quote against this worksheet, send it to admin@sycgeeks.ca or book time. If the live site is down today, skip the worksheet and call.



