How long does it take to build a website?

Ask a web designer how long a website takes and you will usually get a shrug and “it depends.” That answer is technically correct and practically worthless to a business owner trying to decide whether to start today or next month. The honest version names the three paths that actually exist, and the real number attached to each, because the paths are different enough in mechanism that averaging them into one vague estimate hides the decision you are actually making.

Notice what actually separates these three numbers: it is not effort, and it is not quality — it is who is doing the work and how much back-and-forth the process requires. An AI builder like Black Nile compresses the writing, layout, and design decisions into a single generation step, so the “about a minute” number is the entire build, not a rounding trick. A manual builder hands you the tools but not the decisions, so the hours go into you making dozens of small choices yourself. A custom designer inserts a second person’s calendar into your timeline, and most of the weeks are not spent designing — they are spent waiting for drafts, giving feedback, and waiting again.

The rest of this guide breaks down what actually happens inside each of those timeframes, names the specific factors that push a build faster or slower within its path, and is explicit about the one thing none of these numbers tell you: how long it takes your finished site to actually start working for you once it is live.

“About a minute” sounds like marketing shorthand until you see what is actually happening inside it. On Black Nile, you start by signing up with a magic-link email — no password to create, no card to enter. From there, the form asks for exactly what a site needs to be built around: your business name, a category (there are 18 to choose from, covering trades, food, retail, health, professional services, and more), a description of what your business does in your own words, your city, and your phone number. Style and logo are optional — you can hand the AI a direction to work from, or let it decide.

From that handful of inputs, the AI writes every section of the site — headline, services, about, proof, contact — picks a theme suited to your category from a set of purpose-built template families (Ironclad for trades, Cura for health providers, Praxis for professional services, Atelier for storefronts, and others spanning restaurants, real estate, fitness, education, hospitality, and nonprofits), and generates the SEO metadata your page needs to be indexable. That generation step takes 30 to 60 seconds — “about a minute” is the honest, safe way to describe it, because the real number moves slightly with how much the AI has to write.

What happens after generation matters as much as the generation itself. You land on a preview of the finished site, not a blank editor — you are reviewing a complete draft, not building one from scratch. If it is right, you publish with one tap and the site goes live at yourbusiness.blacknile.app. Your first site generation is free; there is no payment step between describing your business and having a published, working site.

The reason this is worth walking through in detail is that the minute number is easy to dismiss as too good to be true, or as describing something thin. It does not describe something thin — it describes a different order of operations. A manual builder makes you supply the content and make the design decisions before you have anything to look at. An AI builder gives you a finished draft first and asks you to react to it. Reacting to a draft is faster than producing one from a blank page, and that ordering, not any shortcut in quality, is where the sixty seconds comes from.

A manual drag-and-drop builder is not slow because the software is clunky. Modern builders are genuinely well made. The hours go somewhere specific and predictable: first, you scroll through a template gallery trying to find one that roughly fits your business, a decision that alone can eat twenty minutes if you are indecisive about which of forty similar-looking “local service” templates to pick. Then you go section by section replacing placeholder text with your own copy — copy you have to write yourself, from a blank cursor, which is the single biggest time sink in the whole process for most small business owners who are not writers.

After the copy comes photos — sourcing your own or picking stock images that do not look obviously stock, resizing and cropping them to fit the template’s frames, and swapping out whatever came pre-loaded in the demo. Then layout adjustments: the template’s default spacing looks fine on the desktop preview and cramped on your phone, so you switch to mobile view, notice three things overlapping, and go back to fix them. This loop — desktop check, mobile check, small fix, recheck — is where a lot of the “few hours” actually accumulates, and it is easy to lose an afternoon to it without producing a site that looks meaningfully more polished than a well-generated one.

None of this makes manual builders a bad choice — plenty of business owners want that level of direct control over every pixel, and for someone who enjoys the process or has very specific design opinions, the hours are well spent. But it is worth being honest about what is consuming the time: it is rarely the software being difficult. It is the fact that you, not an AI and not a designer, are the one making every content and layout decision, one at a time, from scratch.

A custom-designed small business website typically runs four to twelve weeks from kickoff to launch, longer for anything with unusual complexity. The natural assumption is that a person is simply working on your site for weeks straight, but that is not what fills the calendar. A typical project moves through a discovery call and brief, a round of wireframes showing the site’s structure before any visual design happens, one or two rounds of visual drafts, a revision cycle where you give feedback and the designer incorporates it, development to turn the approved design into a working site, and a final QA and launch pass. Each of those stages is a handoff, and handoffs are where time actually disappears.

A designer might spend six focused hours on your actual homepage design. But that design sits in your inbox for four days before you have time to review it, your feedback takes another two days to write up properly, the designer is juggling three other client projects and slots your revision in the following week, and the cycle repeats for the second and third rounds. None of the individual people involved are being slow — the project is slow because it is a conversation stretched across two calendars, and calendars do not compress the way a single person’s focused work does.

What that time buys, when the process works well, is real: a site designed specifically around your brand, your photography, and your exact positioning, with a human thinking through decisions an AI or a template gallery will not make for you — how your specific story should be told, what makes your specific business different from the five competitors down the street. For a business with a distinctive visual identity or a complex positioning problem, that bespoke attention is worth the weeks. For a plumber who needs services, a service area, and a phone number online this week, it is a lot of calendar to spend on a problem that does not require it.

This is the part of the “how long does it take” question that gets glossed over constantly, and it deserves to be stated plainly: how fast you can build a website and how fast that website starts ranking on Google are two completely different clocks, and building fast does not make the second one run any faster. A site generated in a minute and published on the spot is genuinely live the moment you publish it — visitors can reach it, your phone number and contact form work immediately. But Google still has to discover the page, crawl it, index it, and build enough confidence in it relative to competitors before it shows up for a real search like “plumber near me.” That process takes real time — commonly weeks, and for competitive local terms, months of index age and consistent signals before rankings settle into a stable position.

Submitting your new site to Google Search Console can speed up the discovery step from weeks down to days, but discovery is not ranking. A brand-new page, however well written, is competing against pages that have been live and accumulating trust signals — reviews, links, consistent traffic — for months or years. Nothing about a fast build shortcuts that accumulation. The honest way to think about it: publishing fast starts the ranking clock sooner, which is a real advantage over waiting, but it does not fast-forward the clock itself.

This matters because it is easy to walk away from a one-minute build expecting one-minute results, and then feel misled a month later when the site is not yet the top result for your service. It is not evidence the fast build failed — it is evidence that build speed and search performance were never the same variable. The practical takeaway is to treat “how long to build” and “how long to rank” as two separate questions with two separate answers, and to publish as early as reasonably possible precisely because the ranking clock only starts once you do.

Within any of the three paths, the real timeline moves on a handful of factors that have nothing to do with which builder or designer you choose. The first is content readiness: do you already know your services, your service area, and have usable photos, or are you inventing that content on the spot? A business owner who walks in with a clear answer to “what do you do and where” moves through any build path faster than one who is figuring out their own positioning while building the site.

• Page count. One page moves faster than five, on any path. On Black Nile, Free and Starter sites are one page, built and published in about a minute; multi-page sites — up to five pages per site, plus booking and store lead-capture tools for gathering appointment requests and order inquiries as leads you confirm yourself by call or text — start on the Pro plan at $29.99/month, and simply involve more content decisions to make regardless of how fast the underlying tool is.
• Who decides. A solo owner making every call moves faster than a business with a partner, a spouse, or a committee that all need to sign off on the copy and the photos. This factor alone can turn a one-hour manual build into a one-week manual build, independent of the software.
• Revision appetite. How many rounds of changes you want before you are satisfied. On Black Nile, ongoing edits work by typing a request in plain English — “add my Saturday hours” — and the AI drafts the change for you to review and publish, with monthly update slots that scale by plan (Free 1, Starter 10, Pro 50, Growth 200). A business that wants to iterate constantly benefits from more update headroom; one that wants to publish once and rarely touch it does not need much.
• Customization ambition. The more specific and unusual your design requirements, the longer any path takes — this is the one factor that genuinely favors custom design over the other two paths.
• How much you trust the first draft. Reviewing and approving a generated or templated draft is inherently faster than deliberating over infinite blank-page options — the paradox of choice slows manual builds more than most people expect.

Notice that four of these five factors are about the business owner’s own readiness and decision-making, not about the builder. That is worth sitting with: switching tools changes how much of the timeline is friction versus how much is genuine decision-making, but it does not eliminate decision-making. The fastest path is the one that removes the most friction from decisions you have already made, and adds the least overhead to decisions you have not.

It is tempting to assume a one-minute site must be shallower than a twelve-week one, the same way it is tempting to assume anything that took weeks must be better for having taken longer. Neither assumption holds up. A fast AI-generated site can be a fully professional, working business website — real services, a real service area, real contact capture, live and indexable the same day you describe your business. A slow custom project can also fail to deliver, dragging past its estimated weeks with a design that still does not say clearly what the business does. Time spent is not a quality signal in either direction; it is a process signal, telling you how the site was made, not how well it will work.

The honest way to choose is to match the path to the actual job, not to a belief that speed and quality trade off against each other. If the job is: get a real business online, capturing leads, ranking for local search, this week — the AI path is not a compromise, it is the correct tool, and the minute it takes to build is time genuinely not needed for that job. If the job is: solve a genuinely unusual design or brand problem that a template or a description cannot capture — the weeks a custom designer spends are also not wasted, they are buying something the faster paths structurally cannot produce.

Most small businesses reading a guide like this are trying to answer the first question, not the second — they need a working site, not a design statement, and they need it now rather than next quarter. For that business, the right frame is not “is fast good enough” but “does this specific business have a design problem that actually requires weeks to solve.” Most do not. And for the ones that later grow into needing more — more pages, more updates, more sites — the path does not have to restart from zero; it extends from wherever the fast build already got you live.