What a small business website actually needs
Four to six pages, and a short list of the things everybody adds that nobody reads. What to build and what to leave out.
Four to six pages, each answering one question a customer actually has. That is the whole specification for most small businesses, and the reason websites cost more than they should is that almost nobody stops there.
The pages you need
Home
The one page most people will see, and often the only one. It has to answer, within a few seconds: what you do, who for, where, and what to do next.
Put the important thing at the top. Not a full-screen photo with your logo on it — the actual sentence that says what you do. Then a way to contact you that does not require scrolling.
What you do
Your services, in the words customers use rather than the words your industry uses. If you do five things, list five things. If one of them makes most of your money, put it first and give it more room.
Include prices if you can. Publishing them filters out the people who were never going to buy and shortens every conversation with the people who were. "From $X" or "typically $X–$Y" is enough.
Proof
The page that decides it. Photos of real work, real customer comments with real names, whatever you are accredited or insured for. If you have done work for a name people recognise, say so.
This is the page most small businesses skip and the one that converts.
Contact
Phone number, email, address, hours, area covered. A short form for people who would rather type than call. Make the phone number a link that dials when tapped, because most of your visitors are on a phone.
Optional: about
Worth having if trust is the sale — consultants, clinics, anything where people are choosing a person. A real photo and a real paragraph. Skip the stock imagery and the sentence about being passionate about excellence.
Optional: one page for your best-selling service
If most of your income comes from one thing, give it its own page. It is the page that will rank, and it is the page you can send people to.
What to leave out
Nearly every small business site carries these, and none of them earn their place.
A rotating slider on the homepage. People do not wait for slide two. It pushes your actual message down and slows the page.
A blog you will not write. Three posts from two years ago is worse than no blog. It signals abandonment.
A newsletter signup you will never send. Same problem.
Stock photos of people in an office. Everyone can tell. A phone photo of your actual van, shop or workbench beats it every time.
Long history sections. Almost nobody reads how the company began. Two sentences, on the about page, if at all.
A gallery with sixty photos. Eight good ones beats sixty average ones. Nobody reaches the end.
Social feeds embedded in the page. They load slowly and send people away from the page where they were about to call you.
The things nobody mentions until later
These are not pages, but leaving them out is what makes a website quietly bad.
- It has to be fast on a phone on a poor connection. Most of your visitors are on mobile data. A site that takes six seconds to appear has already lost a share of them.
- Your business details need to be marked up properly so search engines can read your address, hours and phone number. This is how you turn up in local results.
- Every page needs its own description. It is what people see in search results, and it decides whether they click.
- It must work with a keyboard and a screen reader. It is not difficult if it is done from the start, and it is expensive to bolt on afterwards.
- Someone has to be able to change it. Hours change. Prices change. If editing a line of text is a purchase decision, the site will drift out of date and become actively misleading.
The short version
Four to six pages. Real photos. Real prices if you can. Fast on a phone. Easy to keep accurate.
That is a complete small business website, and it should not take months or cost thousands. Ours is $499 to build and $49 a month to run, and the monthly fee exists partly so that changing a line of text never has to be a decision.