Most guides to building a business website answer the wrong question. They tell you how to pick a template, choose a domain and write an About page — and you end up with a brochure. A nice-looking page that lists what you sell, gives a phone number, and quietly does nothing for the rest of its life.
The question worth answering first is not how do I build a website. It is what do I want this website to do when somebody lands on it at 9pm on a Sunday?
There are only really three honest answers, and they lead to three different builds.
Step 1: Decide which of the three you actually need
A presence page. One page. Who you are, what you sell, where you are, how to reach you. It exists so that when somebody searches your business name, something you control comes up instead of nothing. If you are a service business that closes every deal on a call, this is sometimes genuinely all you need.
A catalogue. Your products, with prices, that people can browse and then message you to buy. This is where most small retailers land by accident — usually as an Instagram grid or a WhatsApp catalogue — and it is a real step up from nothing. The limit is that every order still passes through a human conversation, so your evenings belong to your customers.
A store that takes the order. Browse, add to basket, pay, done. The order arrives as an order, not as a message you have to read, price and confirm.
Be honest about which one you want, because the third is the only one that keeps working when you are asleep, and the first two are much cheaper. There is no shame in starting at a presence page. There is a real cost in building the third one badly.
Step 2: Understand the trap before you pick a tool
Here is the failure almost nobody warns you about, and it has nothing to do with web design.
You already have a system for knowing what you have in stock. It might be a counter book, a spreadsheet, a till, or your own memory. When you add a website that also holds your products and prices, you now have two of them. Two places listing what you sell. Two places that think they know how many are left.
They will not agree. They start out in step on launch day, and they drift from the first sale onward. Someone buys the last pair of shoes in the shop; the website keeps selling them for another four days. You run a price change at the counter; the website quotes last month's price to a customer who screenshots it.
The usual advice is to fix this with an integration — a plugin that syncs your store to your till on a schedule. That works right up until it does not, and when a sync job fails silently you find out from an angry customer rather than from the software.
The way out is not a better sync. It is not having two lists in the first place. One catalogue, one stock count, read by both the counter and the website. An item sold at the till is gone from the website because it is the same number, not because a job copied it across.
This is the single most important decision in the whole exercise, and it is one you make when you choose the tool — not something you can retrofit later without redoing the catalogue.
Step 3: What it actually costs
Three honest cost lines, and one that gets hidden.
The domain. Roughly $10–20 a year for a .com, and local domains are usually in the same range. Buy it in your own name, from a registrar you can log into. This matters more than it sounds: a surprising number of small businesses discover years later that their domain is registered to a developer who has stopped answering.
The build. Anything from nothing — if you use a builder and do it yourself over a weekend — to several thousand dollars for a custom agency site. A templated store you configure yourself is the sensible middle for most retailers, and it is the option that gets you live this month rather than next quarter.
The running cost. A monthly subscription for the platform, plus payment processing fees on each sale. Watch for the second subscription: with most setups, the point of sale and the online store are two separate products with two separate bills.
The hidden one: entering your catalogue. This is the cost nobody quotes, and it is the one that kills projects. Every product needs a name, a price, a category, a photograph and an opening stock figure. A hundred products is a long evening. A thousand is a fortnight you do not have, which is why so many business websites launch with eleven products on them and never grow.
We wrote the full arithmetic up in what a store costs to build, including the numbers most quotes leave out.
Step 4: Get it live
Assuming you have chosen the third option — a site that actually takes orders — the order of work is:
- Secure the domain. In your own name. Do this first; everything else can be changed later, and this is the part that is painful to undo.
- Load the catalogue once. Names, prices, categories, photos, opening stock. If you have an existing price list, a spreadsheet or even a photograph of a handwritten list, you should not be retyping it — see the next section.
- Connect the money. Mobile money and cards, into your own business account. Test it with one real transaction of a few shillings before you tell anyone the site exists. Do not skip this; "it looked like it worked" is not the same as money arriving.
- Set delivery and collection. What you charge, where you cover, and whether customers can collect in person. Vague delivery terms generate more messages than the orders are worth.
- Write five product descriptions properly. Not all of them — your five best sellers. Those are the pages people will actually land on.
- Make one test purchase yourself, end to end. Buy something from your own shop, as a stranger would. You will find at least one thing wrong, and it is better that you find it than a customer.
- Then tell people. Not before step 6.
Step 5: Do not type the catalogue in by hand
This deserves its own section because it is the step at which most business websites quietly die.
You do not need to retype your products. If you have them in a spreadsheet, import the spreadsheet. If you have a supplier invoice or a printed price list, that is a file too. If what you have is a photograph of a handwritten page, that is still a starting point — Amina, the assistant built into sell.ke, will read it and build the catalogue from it.
And if you would rather not do it at all, we will load it for you. That is a paid, done-for-you setup — priced and described on the page, with a turnaround stated up front — not a free extra. We mention it plainly because the most common reason a store never launches is an owner staring at an empty product list on a Tuesday night.
What "good" looks like six months later
A business website is working if:
- Orders arrive as orders, with the money already taken.
- The stock figure on the site is the stock figure in the shop, because they are the same figure.
- You can change a price in one place.
- Month-end does not involve reconciling the website against the till by hand.
- You can see, without asking anyone, what sold today and what it made you.
If you are doing any of those twice — once for the shop, once for the site — the setup is costing you time every single day, and that cost compounds quietly.
Frequently asked questions
Do I need a website if I already sell on Instagram and WhatsApp? Those channels are rented. They are excellent for reach and terrible for ownership: you do not control the algorithm, the account, or whether the platform still exists in your market in three years. A site on your own domain is the asset; the social channels drive people to it. Most businesses should run both, with the catalogue living in one place and the social posts pointing at it.
How long does it take? If the catalogue is already in a spreadsheet, a working store is a day's work. If you are entering products one at a time by hand, it takes as long as the catalogue takes — which is the real answer, and why step 5 matters more than the design.
Can I start with a presence page and add selling later? Yes, and it is often sensible. The thing to avoid is building the presence page on a tool that cannot grow into a store, because then "adding selling later" means starting again — including the catalogue.
What about SEO? Have real pages for real products with real descriptions, a fast site, and a clear structure. That is most of it for a small business. The sophisticated work only pays off once the basics are in place, and the basics are mostly just "the information is actually on the page".
Ready to put one together? Start a free 7-day trial — no card needed — or book a free demo and we will walk through your product list with you. If you want to read more first: how to set up an online store, and what a store costs to build.