The BVM PulseOpinion & Thought Leadership

    Why most SME websites do not earn

    A business pays for a website, the website goes live, and nothing changes. This is common enough that many owners conclude websites do not work here. The sites are not failing at being websites. They are failing at being useful.

    Collins Lartey31 July 2026Opinion & Thought Leadership
    Share this
    Share

    Founder of Bold Vision MultiTech. Building digital and AI infrastructure for African businesses, learners and institutions since 2016.

    A pattern repeats often enough in this market to be worth naming. A business decides it needs a website. It pays somewhere between GHS 2,000 and GHS 15,000. Something goes live: a homepage, an About page, a Services page, a contact form. Everyone is pleased for about two weeks.

    Then nothing happens. No enquiries arrive through it. Customers still call the number they already had. A year later the domain renewal notice arrives and the owner has to think hard about whether to pay it.

    The conclusion many owners draw is that websites do not work in Ghana. That conclusion is wrong, but it is reasonable, because the thing they were sold genuinely did not work.

    The brochure problem

    Most SME websites are built as brochures. They describe the business: who we are, what we do, our values, our team, contact us. Every page is about the company.

    The trouble is that nobody wakes up wanting to read about a company. People arrive with a task. They want to know whether you sell the specific thing they need, what it costs, whether you deliver to their area, whether you are open, and whether anyone will actually reply. A brochure answers none of these, so the visitor leaves and calls someone who will answer on WhatsApp.

    A website earns when it does a job the customer already wanted done. It fails when it explains a company nobody asked about.

    This is why "we need a website" is usually the wrong starting question. The right one is: what do people ask us before they buy, and can the site answer that faster than a phone call?

    Four things that separate a site that earns

    1. It answers the buying questions, in public

    Price, delivery, location, availability, payment methods, turnaround time. Businesses hide these constantly, usually reasoning that a conversation lets them sell better. In practice, an unanswered price question does not become a conversation. It becomes a visitor who leaves.

    Publishing this information does something else too. It is the raw material for being found. A page that plainly states what you sell, where, and at what cost is a page a search engine can rank and an AI assistant can quote. Vague positioning language is invisible to both.

    2. It loads on a real phone on a real network

    77 percent of Ghanaian web traffic arrives on a mobile device, frequently on a metered connection with inconsistent signal. A site that takes eight seconds to appear has lost most of its visitors before they see anything at all, and the owner will never know, because those people never register as traffic worth reporting.

    We hold our builds to loading in under three seconds on a 3G connection. That is not a stretch goal, it is the baseline this market actually runs on. A beautiful site that only performs on office broadband is a site built for the wrong country.

    3. It connects to how the business really operates

    If every enquiry ends with "call us" and the phone is answered inconsistently, the site is a dead end with good typography. A working site routes the visitor into the channel the business is actually responsive on, which in Ghana usually means WhatsApp, and captures who they were so they can be followed up.

    4. It is built to be found, including by machines

    More than 60 percent of searches now end without a click, because the answer appears on the results page or comes from an assistant that read the web on the user's behalf. A site with no structured data has no way to participate in that. It can be crawled, but it cannot be quoted accurately.

    Getting this right is unglamorous: clear factual statements, structured data describing the business and its products, a genuine FAQ, consistency between every place the business appears. This is the discipline we call Prompt Engine Optimization, and it is increasingly the difference between existing online and being recommended.

    The uncomfortable part

    Almost none of the above is design work, and that is why so many websites disappoint the people who commissioned them.

    A business asks for a website and receives a website: pages, layout, colours, images, all delivered competently. What it needed was for a stranger to find it, trust it, get an answer, and reach a human. Those are different briefs, and only one of them changes revenue. A site can be attractive and still fail every one of the four tests above.

    This is also why price comparison on websites is misleading. Two quotes for "a five page website" can describe completely different things: one is a layout, the other is a system for being found and converting. They cost differently because they are not the same product.

    How to judge your own site

    Four questions, and you can answer all of them yourself in ten minutes.

    1. Open it on your phone on mobile data, not office wifi. Count the seconds until you can read something. If you are past three, most visitors are already gone.
    2. Search for what you sell plus your city, as a customer would type it. If you are nowhere, the site is not doing the one job that brings strangers.
    3. Ask someone outside the business to find your price and delivery terms. Time them. If they cannot, neither can a customer.
    4. Ask an AI assistant about your category in your city. If it does not know you exist, or describes you wrongly, your information is not structured well enough to be cited.

    A site that passes those four is worth paying for every year. A site that fails them is a recurring cost with no return, and no amount of redesign fixes it, because the problem was never how it looked.

    Build what Africa needs next.

    Partner with BVM to scale a platform, transform an institution or build a new digital system.