What actually happens week by week
Contractors ask this question because every quote they've gotten has a different number attached to it, and none of them explain what's actually inside that number. A quote of "2 weeks" and a quote of "6 weeks" might be describing two completely different jobs. Here's the build broken into the phases that actually happen, in order, on a hand-coded site (no page builder, no theme to wrestle, no plugin updates to babysit later, no monthly software fee tacked onto the build cost).
| Phase | What happens | Typical length |
|---|---|---|
| Kickoff & content gather | Trade angle, service list, service areas, licensing, photos, past job details | 2-5 days |
| Structure & copy | Page map built, home + trade pages + service-area pages written | 1-2 weeks |
| Build | Hand-coded HTML/CSS, forms wired, mobile pass, speed pass | 1-2 weeks |
| Review & revisions | Owner reads every page, requests changes, we turn them | 3-7 days |
| Launch & DNS | Domain pointed, site goes live, forms tested live | 1-2 days |
Add those up and you land at 3 to 6 weeks for a site with a handful of trade pages and a real service area. A one-trade, one-location site with no blog and a short service list can clear the whole process in closer to 3 weeks. A multi-trade outfit covering six counties with a dozen service pages, each one written to actually mention that county by name, runs closer to 6.
Notice that the build phase itself, the part most contractors picture when they hear "building a website," is usually the shortest stretch on the calendar. Writing the copy that makes each page worth indexing takes longer than coding the page that holds it. That's backwards from what most owners expect, and it's exactly why timeline quotes that skip the copy phase come in so much shorter: they're quoting the part of the job that takes the least time.
None of this includes ongoing SEO ranking work, map pack work, or an AI-search visibility campaign. Those are separate, ongoing engagements that start after launch, not phases of the build itself. A contractor asking "how long until I rank" is asking a different question than "how long until my site exists," and conflating the two is where a lot of frustration with web shops starts.
Why template sites go live faster (and what that speed costs you)
A $99/month DIY builder or a stock template from a marketplace can be "live" in an afternoon. Pick a layout, drop in a logo, type your phone number into a header field, hit publish. That speed is real, and for some businesses that's genuinely all they need. What's also real is what you're looking at when it's done: a design a hundred other contractors in a hundred other towns are also running, stock photography of someone else's crew on someone else's job site, and placeholder copy ("Welcome to our company, we do quality work") that never gets replaced because nobody budgeted time to write it.
Fast template sites skip the two phases that actually move the phone: written trade pages that answer what a homeowner is actually searching for, and service-area pages that tell Google and AI search engines where the business actually works. Skip those and the site loads on day one and sits invisible for the next two years, because there's nothing on it specific enough for a search engine or an AI answer engine to point to.
We've seen the pattern often enough to say it plainly: the DIY build isn't slow because it's careful, it's fast because it's empty. A page builder can spin up twelve pages in an hour, but twelve empty pages don't rank any better than one. Filling that emptiness later, on a platform built to keep the business locked into a monthly template fee and a walled-off editor, usually costs more in time and money over two years than doing it right the first time would have.
- Template site: live in 1-3 days, but pages are generic, thin, and duplicated across thousands of other small business sites on the same platform
- Custom hand-coded site: live in 3-6 weeks, but every page is written for the specific trade and the specific towns served
A slower launch that gets quoted by name in an AI search answer beats a fast launch nobody reads past the homepage. Speed to launch and speed to results are two different clocks, and a lot of contractors only find out they were sold the wrong one after a year of silence from the phone.
What slows the timeline down (and it's rarely us)
The single biggest variable in any contractor website timeline isn't code, it's content turnaround. We can write and build fast on our end because that's the part we control every day. What we don't control: your license number, your actual service radius, or captions for job photos we don't have yet. Here's what reliably stretches a 4-week build into 8 or longer:
- Photo delay. No job photos on hand means either stock imagery placeholders that sit there until real photos show up, or a multi-week wait while the contractor digs through a phone camera roll or asks a foreman to send over shots from last month's jobs.
- Slow sign-off. One review round turned around in 2 days keeps the whole build moving. One review round that sits in an owner's inbox for 3 weeks because the season got busy adds 3 weeks to the calendar, flat, no matter how fast the build phase itself went.
- Scope creep mid-build. Adding four more service-area towns or a second trade after the page map is already locked doesn't just add those pages, it resets the copy phase for consistency across the whole site.
- Licensing or insurance paperwork. Some states and some trades require license numbers or specific disclosure language displayed on the site. Waiting on a renewed certificate or a state license lookup holds up launch even when every page is otherwise ready.
- Domain and DNS access. If nobody currently has the login to the domain registrar (a shockingly common problem when a domain was registered by an employee who left, or a previous web person who's no longer answering calls), launch day slips until that access gets tracked down or the domain gets transferred.
The fix for all five is the same: front-load the content gather before the copy phase starts, not during it. The contractors who hand over photos, a service list, and a licensing packet in the first week are consistently the ones who launch in week 4, not the ones still chasing down a DNS password in week 9.
Does the timeline change by trade?
Some. The build mechanics don't change (a roofer's site and an electrician's site get built the same way, on the same hand-coded stack, wired to the same forms), but the content volume does, and content volume is what takes time to write, review, and get right.
| Trade pattern | What adds time |
|---|---|
| Storm-driven trades (roofing, gutters, exterior) | Insurance-claim language needs to be accurate and non-generic; often more service-area towns to cover because storm damage isn't confined to one zip code |
| Licensed trades (electrical, plumbing, HVAC) | License numbers, permit language, and the distinction between a service call and a full install need to be correct and consistent across every page they appear on |
| Multi-crew outfits (general contractors, remodelers) | More job types means more trade pages, which means more copy to write, more photos to source, and more review rounds |
| Single-service specialists (gutter guards, fencing, one-trade shops) | Fastest builds: one core service, a tighter page map, and less back-and-forth because there's less ground to review |
A specialist with one core service and a tight service area is closer to the 3-week end of the range. A multi-trade outfit with a wide territory and a dozen job types, each deserving its own page instead of being crammed into a single generic "services" page, is closer to 6. Neither timeline is wrong: it's proportional to how much ground the site actually has to cover to be useful to the homeowners searching for it.
Trade also affects what has to be true on the page before it can go live. A plumber's site claiming licensed and insured work needs the license number to back it up. A roofer's site mentioning storm damage or insurance claims needs to be careful not to imply the business handles the claim itself if it doesn't. These aren't timeline killers on their own, but they're one more reason the content-gather phase in week one matters more than the coding phase in week three.
Can you rush it? What actually gets cut
Yes, but a rush changes what's on the site, not just when it goes live. If a contractor needs something live in under two weeks (a competitor just launched a new site, a slow season is ending and calls need to start now, a trade show or home show is coming up fast), the honest options are:
- Launch narrow, expand after. Home page, one or two core trade pages, and the top 2-3 service-area towns go live fast. The rest of the trade pages and towns get built and added in the following weeks, once the core site is already live and taking calls.
- Reuse what exists. If photos, a service list, and licensing info are already gathered in one place (not scattered across three group texts and a filing cabinet), the content-gather phase shrinks from a week down to a day, sometimes less.
- Compress the review round. A same-day review turnaround instead of a week-long one saves a full week off the calendar, no build shortcuts required, just faster decisions on the owner's end.
What doesn't get cut under a rush: hand-coded markup, page speed, mobile behavior, and working forms. Those aren't extra time built into the schedule, they're the baseline every site gets regardless of how fast it needs to move. A rushed site from us still loads under 2 seconds and still has a form that reaches a real inbox with a honeypot behind it to keep the spam out. What gets thinner under a rush is page count and copy depth, not the build quality underneath what does ship.
Rushing a 6-week job into 2 weeks by cutting corners on the code instead of the page count is how a contractor ends up back at this same question eight months later, looking for someone to rebuild what broke or what never worked right in the first place. A narrower launch that's built solid beats a full-scope launch that's built thin. If the calendar is truly non-negotiable, say so in the first call: a phased launch plan gets drawn up around the hard date instead of the hard date getting missed by a build that tried to do everything at once.
What we need from you to hit the fast end
The contractors who launch closest to the 3-week end of the range all did the same handful of things in week one, before the build even started. This is the punch list, and it's short on purpose:
- A list of every trade or service the business actually performs, not an aspirational list of what the owner wishes the crew did, the real current one
- Every town, county, or zip code served, ideally ranked in order of priority so the highest-value areas get built first if a phased launch makes sense
- License numbers, insurance info, and any state-required disclosures that need to appear on the site
- Job photos: even rough phone photos of finished work beat polished stock images of somebody else's crew every time, because a homeowner can tell the difference
- One person designated to approve copy and give the go or no-go on launch, not a rotating committee that each wants a different word changed
Hand that over in the first meeting and the copy phase starts immediately instead of stalling on a follow-up email that sits unanswered for a week and a half. That single change, having the content ready before the clock starts instead of gathering it during the build, is worth more to the timeline than any amount of pushing on our end could ever be.
It's also worth saying what this timeline does not include, because scope confusion here causes more frustration than the actual wait. Everything past launch (climbing rankings for competitive keywords, winning the map pack against local competitors, getting quoted by name in ChatGPT or Google's AI Overviews) is a separate, ongoing track with its own timeline measured in months, not weeks. The build timeline covers the asset itself: the site that goes live, loads fast, and is structured so those later efforts have something solid to work with. What happens to its visibility over the following months is its own conversation, with its own separate scope, its own separate contract, and its own separate set of expectations about what "progress" looks like month to month.