You need a site, so you ask what launching one involves, and the answer comes back as a list: sitemap, wireframes, a comparison of four platforms, personas, a content plan, an analytics stack, an SEO audit. Six weeks later, the list is longer, and nothing is live. Launching a first site means deciding four things: the domain, the platform, the one page that carries the argument, and the one form. Every other decision on that list is why a project that should take three weeks takes four months.
The domain is the only one of the four that is expensive to undo.
Founders usually have the effort backward. The platform gets a month of comparison. The domain gets ten minutes on a registrar at midnight, because it feels like a formality.
It is not a formality. Once a domain is on the deck, in everyone’s email address, on the invoices and inside the contracts, changing it is a project with its own timeline. The other three decisions you can redo in a week.
The test for a domain is not whether it is clever. It is whether someone who heard it once on a call can type it correctly the next morning without asking you. That rules out most of what a naming exercise produces: deliberate misspellings, extensions people type as .com anyway, and hyphens that have to be explained every time.
Take the name you can say out loud, buy the plainest version of it you can get, and stop. If the name itself is still open, that is a branding decision rather than a website one, and it is worth finishing before the site starts. An afternoon is the right amount of time to spend here. A month is not.
The platform matters less than the person who will edit it.
Founders treat this as a technical question. It is a staffing one. The right platform is the one where whoever needs to change a sentence on a Tuesday can change it without opening a ticket. At this stage, that person is a founder.
If the site needs an engineer’s afternoon to fix a headline, the headline stops getting fixed, and a site whose copy has not changed since launch is a site nobody is learning anything from.
The same reasoning rules out building it yourself. An engineer’s week is the most expensive thing an early company owns, and spending it on a four-page marketing site buys nothing a hosted builder already does.
Two conditions on whatever you pick. You own the domain and can point it somewhere else, and the content comes back out in a form you can move. Everything else about the choice is reversible at this size, which is why it does not deserve the month it usually gets.
One page has to carry the argument, and writing it is the hard part.
First sites tend to have a handful of pages, and one of them gets read. Build that one well and keep the rest short.
The home page answers three things for someone who arrived with a reason to look: what you make, who it is for, and what happens next. Anything on the page that serves none of the three comes off.
The common failure is writing the site for the round. A first site written for investors reads like a deck, and the customer did not come for a market-size slide. Write it for the customer. Candidates and investors will read a site written for customers and understand it. It doesn’t work the other way around.
This is the section that stalls, and it stalls because it is not a website problem. It is a positioning problem that arrived looking like one. The first sites we get called into that stall were seldom stuck on the build; they were stuck on a sentence nobody had written, or on pages and forms no one mapped and built. No platform will produce that sentence for you, and no amount of design will cover for it being missing.
One form, and someone has to check that it arrives.
The site needs exactly one way to make contact, and it belongs on every page. Not a chat widget nobody is staffing. Not a calendar link that assumes the reader is ready to book. Not an address in the footer pointing at an inbox three people half-read.
Ask for as little as you can act on. A name, a way to reply, and a line about what they are building is enough at this stage. Every field past that is another reason to close the tab, and a qualifying question asked before a conversation is a question asked of a stranger.
Then test it the way a stranger would. Submit it from a phone, on cell data, not logged in, and confirm that the message reaches a person and that a reply goes back the same day. This is the quietest way a launched site fails: the form posts, the page says thank you, and the mail never leaves the server. Nobody finds out until someone mentions, months later, that they tried to get in touch.
Everything else waits until the site has done something.
The list you were handed is not wrong. It is early. A blog with nothing behind it, a pricing page for pricing you are still testing, case studies you don’t have yet, a cookie banner for analytics you haven’t installed. Each is defensible on its own, and together they are a launch date that keeps moving.
Ship at three weeks with four pages and change it weekly. A first site is not an asset you build once. It is the current version of your argument, and the argument moves.
The site is also not the marketing. It is where the marketing lands, which is why it comes after positioning and before the channels. Launching it is not a moment worth announcing. It is the point at which the rest of the work has somewhere to point.
Spend an afternoon on the domain, most of the three weeks on one page, and the last hour submitting the form from a phone before anyone links to the site. Brand, Positioning, and Website is the order the work goes in, and the site gets easier once the first two are settled.
Tell us what you’re building.