Most people who begin a website project already know where to start and simply cannot say it in writing. They can describe the business to a friend in ninety seconds, and the same content, faced with a homepage, comes out as “professional, modern, and user-friendly.”
That gap is not a writing problem. It is an undecided-ness problem, and it is exactly what planning work is for.
Start from the conversation, not the design
Do not open a design reference. Do not look at competitors yet. Answer these in order, in a notebook or a document, in complete sentences:
- What do you actually sell?
- Who buys it?
- How do people usually find you today?
- What do people ask before they commit?
- What makes you different, stated without adjectives?
- What should a visitor do after reading your homepage?
- What would make this website obviously successful, one year from now?
Question five is where people stall. If the honest answer is “nothing, we are quite similar to others,” that is useful information and it changes what the site has to do: credibility and clarity carry the decision, not differentiation theatre.
Question seven is where the project gets scoped. “Ten good enquiries a month” and “people can find our prices” are different websites.
Then describe the visitor, not the company
Write one paragraph about the person who arrives with no context. Not your ideal client profile — a real one. What happened this morning that made them search. What they already understand, what they do not, and what they are worried about.
A plumber’s customer is worried about being overcharged. A review centre’s parent is worried their child will fall behind. A buyer comparing POS systems is worried about picking the wrong one and having to redo it.
Those worries, answered on the page, are what makes a website feel relevant. Nothing about your company is the interesting part until trust is established.
Now the pages can be named
Take the answers above and turn them into a list, with a job attached to each line:
- Home — states the offer and routes people onward.
- Service or offer pages — one per thing you sell.
- About — why a stranger should trust you.
- Proof — gallery, projects, classes, catalog, reviews: whatever your trade uses as evidence.
- Contact — how to reach you and what happens next.
Add a page only when you can name the question it answers. That rule is the whole of information architecture for a small site, and it keeps a five-page project from becoming a twelve-page one.
The three structures that cover most cases
One clear offer. Home, one offer page, proof, contact. Short, and usually the right answer for freelancers and single-service businesses.
Several distinct offers. Home with a clear split, one page per offer, one about, one contact. This is the shape for a review centre with three pathways, or an academy with tiered classes.
A catalog. Home, category pages, product or package pages, contact or enquiry. This is what a hardware supplier or a package-based service needs, because the visitor arrives already knowing what they are looking for.
Choosing among these three is twenty minutes of work and determines everything downstream: navigation, page count, platform, and price.
Where planning usually goes wrong
Starting with the platform. “Should we use WordPress?” is a question that has an answer only after the content and update frequency are known. Ask it second, never first.
Migrating undecided-ness. If the offer is fuzzy, the site will be built around a fuzzy centre and every design decision will re-open the argument. Settle the offer, then build.
Collecting opinions instead of information. Five people giving taste feedback on a homepage is not evidence. Ask them the five-second clarity question instead and record what they say.
Waiting for certainty. You will not know which service page matters most until the site is live. Plan for the site to be editable, launch the small version, and let enquiries teach you.
What “done planning” looks like
A short document, usually two to four pages, containing:
- The offer sentence and the audience priority.
- The page list, each with its job in one line.
- The navigation that falls out of that list.
- The content each page needs, and who supplies it.
- The main action on each page.
- The platform decision, with the reason.
- What is explicitly not being built now.
That document is the deliverable. It is what makes a build cheap and predictable, and if you decide to have someone else build it, it is the most useful thing you can hand them.
A reasonable first step, today
Write the seven answers from the top of this page. Then write the paragraph about the visitor who arrived with no context. Then list the pages, each with its job.
If you get through those three and the project still feels large, the missing thing is not more research. It is someone who has done it before telling you which decisions actually matter — which is the entire content of a planning conversation.
Want the structure settled before anything gets built? Discuss your website structure, or send the seven answers through the contact form.