What a business website is actually for

Ask most business owners why they have a website and you'll get some version of the same answer: because you're supposed to. It sits somewhere near "get a logo" and "print business cards" on the list of things a real business does. Once it's live, it's done. Nobody revisits it unless something looks broken.

That answer explains why so many business websites exist and so few of them actually work. A website built to satisfy a checkbox looks like a checkbox: a home page, an about page, a contact page, some stock photography, a paragraph about the company's "passion for excellence." It sits there. It doesn't do anything, because nobody ever asked it to.

This post is about the alternative. Not a longer checklist, but a different starting question. Instead of "what should a website have," ask "what is this website actually for." Once that's answered honestly, the pages, the structure, and the priorities mostly follow on their own.

The problem: a website that doesn't do anything

Here's what an unexamined business website usually looks like in practice.

It loads slowly, especially on a phone on an ordinary mobile connection, because nobody thought about size or speed when it was built. It talks about the business in the abstract, mission statements and stock phrases, without answering the specific questions a visitor actually has: what do you do, what does it cost, are you the right fit for my problem, how do I reach you. The contact information is a phone number in tiny text in the footer, or a contact form that quietly emails an inbox nobody checks. There's no way to tell whether any of it is working, because nobody set up analytics, so the site is a black box that occasionally, maybe, sends an enquiry.

None of this is because the business doesn't care. It's because the website was treated as a one-time purchase, like a sign above a shop door, rather than as something that does a job every day. A sign just needs to exist. A website that's supposed to bring in business needs to work.

The cost of this is quiet and constant, which is exactly what makes it easy to ignore. A slow page loses a fraction of visitors before it even finishes loading, and the scale of that is well documented: the likelihood of a mobile visitor bouncing rises 123 percent as load time goes from one second to ten seconds, according to Google/SOASTA research cited by Think with Google. A page without a clear next step loses visitors who were interested but didn't know what to do. A page that doesn't answer the question someone searched for never gets found in the first place. None of these show up as a dramatic failure. They show up as a slow trickle of enquiries that never quite matches how good the business actually is in person.

It's also worth naming what usually happens instead. A business without a working website doesn't stop getting enquiries, it just gets them through slower, less scalable channels: a referral who happened to have the owner's personal number, a walk-in who found the shop by accident, a WhatsApp forward that reached the right person by luck. Those channels work, but they don't grow the way a findable, functioning website can, and they put a hard ceiling on how many people can discover the business in the first place. The website isn't competing with those channels. Done properly, it's what lets them scale past word of mouth.

Why so many business websites end up this way

Part of it is history. For a long time, "get a website" and "get a brochure printed" were treated as the same kind of task: hire someone, describe the business, get a finished thing, move on. That mental model has stuck around long after the web itself changed into something people use to search, compare, and decide, not just browse.

Part of it is how websites get sold. A lot of agencies and freelance designers compete on how a site looks in a screenshot, because that's what's easy to show in a portfolio. Visual polish is genuinely worth having, but it's not the same thing as a site that converts a visitor into an enquiry. A beautiful site that loads in eight seconds and buries the phone number is still a beautiful site. It's also still failing at its job.

And part of it is DIY (do it yourself) website builders, which solve a real problem, getting something online without hiring anyone, but nudge every user toward the same generic template with a hero image, three feature boxes, and a testimonials carousel, regardless of what that particular business actually needs its visitors to do. The builder optimises for "looks like a website." It has no opinion on whether it works like one.

The pattern underneath all three is the same. Nobody stopped to ask what specific job this website needs to do, for this specific business, for this specific visitor. Once you ask that question, most of the checkbox thinking falls away on its own.

A practical framework: a website is a system with jobs to do

Think of a business website less like a brochure and more like a very patient employee who's on duty around the clock, never has a bad day, and only ever says exactly what you've prepared them to say. Given that, what would you actually want them doing? Broadly, five jobs.

Job 1: Answer the questions before they're asked

By the time someone lands on your website, they usually already have questions in mind: what do you do, what does it cost, do you serve my area, are you legitimate, is this the right fit for my problem. A website's first job is answering those honestly and quickly, before the visitor has to go looking or, more likely, gives up and leaves.

This sounds obvious, but it's the single most commonly skipped job. Plenty of sites talk about the business in the abstract without ever plainly stating what it costs, who it's for, or what happens after someone gets in touch. If a visitor has to guess, most of them won't bother asking. They'll just leave and try the next search result.

Job 2: Prove you're worth trusting

Trust is the quiet gatekeeper of every enquiry. Someone might like what you offer and still not contact you, because nothing on the page told them you're a real, reliable business rather than a shell. This is where genuine social proof matters: real client names where you have permission to use them, real examples of past work, clear information about who runs the business and where it's based. We've written elsewhere about why this has to be genuine: fabricated reviews or vague, unverifiable claims tend to do more damage than having no proof at all, because a sharp visitor can usually tell the difference, and once they suspect one thing is fake, they stop trusting everything else on the page too.

If you're early and don't have client stories yet, say less rather than invent more. A clear, honest description of what you do and how you work is more persuasive than a testimonials section that reads like it was written by the business itself.

Job 3: Make starting a conversation the easiest thing on the page

This is the job most sites quietly fail at. A visitor who's ready to get in touch should never have to hunt for how. That means a phone number or WhatsApp link that's actually visible, not squeezed into six-point footer text. It means a contact form that's short, that actually reaches a real inbox, and that tells the visitor what happens next once they've submitted it. It means the path from "I'm interested" to "I've made contact" is one or two clicks, not a maze.

Every extra field on a form, every extra click to find the phone number, every moment of "wait, how do I actually reach these people" is a chance for an interested visitor to close the tab instead. The businesses that get the most out of their websites tend to be the ones that treat this path as the most important thing on the entire site, because it is.

For most businesses operating locally, that path should end on WhatsApp, not a contact form. A visitor who's already comfortable messaging on WhatsApp for everything else is far more likely to actually send that first message if the site meets them there instead of filling in a form and waiting for an email reply. That only works, though, if the WhatsApp on the other end is actually organised enough to answer. A website that funnels enquiries into a WhatsApp nobody is managing hasn't fixed the problem, it's just moved it one step downstream. We wrote a whole guide on that half of the equation: WhatsApp is your front desk.

Job 4: Be findable for the searches that matter

None of the above helps if nobody finds the page in the first place. A website needs a basic structure that search engines and AI answer engines can actually read: clear page titles, sensible headings, fast load times, and content that plainly answers the questions people search for, in their words, not in internal company jargon. This is less about chasing rankings for competitive, generic terms and more about being the obvious answer for the specific, practical questions your actual customers type into a search bar, things like "web designer in Nairobi" or "how much does a business website cost in Kenya."

This is also where a website and a content strategy meet. A site with strong bones but nothing to say will still struggle to be found. A content system, not just a calendar, is what keeps a site's visibility growing over time instead of flatlining after launch.

Job 5: Show you what's actually happening

The last job is the one that makes the other four improvable: analytics. Without it, you're guessing whether the site works. With it, you can see which pages people actually land on, where they drop off, and whether the contact path is being used. That's what turns "we redesigned the website" from a one-off event into an ongoing process of noticing what isn't working and fixing it, rather than a fresh coat of paint every few years based on nothing more than a hunch that it's time.

The five jobs at a glance

What this looks like in practice

A few illustrative examples of how these jobs play out for different kinds of businesses. As with anything we describe this way, these are illustrations of the framework, not results we're claiming for any specific client.

A small accounting firm might have a beautifully designed site that never says what services cost or how a new client actually gets started. Applying Job 1 and Job 3 here might mean adding a plain-language services page with indicative pricing ranges and a short "how it works" section, plus a booking link that's visible from every page, not just the contact page. The redesign isn't really about how it looks. It's about closing the gap between "visitor is interested" and "visitor has actually booked a call."

A construction or fit-out company might have strong past work but no organised way to show it. Job 2 here is a proper project gallery with real photos and short, specific descriptions, not a vague "our work" page with three stock images. Paired with Job 4, that same gallery content, described in the customer's language rather than industry terms, is also what gets the site found by people searching for exactly that kind of project.

A clinic or salon might get plenty of traffic but very few actual bookings from the site, because the booking process buries itself three pages deep behind a "contact us" link. Job 3 here is blunt: put the booking link, or a WhatsApp button, in the same visible spot on every single page, and cut the number of steps between "I'm on the site" and "I have an appointment" down to the minimum.

Notice that in each case, the fix isn't "make it prettier." It's closing a specific gap between what a visitor needs from the site and what the site actually gives them.

Common mistakes to avoid

Treating launch as the finish line. A website that's never touched again slowly drifts out of date: prices change, services change, and nobody updates the copy. Budget for small, regular updates, not just a one-time build.

Prioritising design over function. A striking design that hides the contact information or takes eight seconds to load on a phone is a design failure, whatever it looks like in a screenshot on a laptop with fast wifi.

Writing for yourself, not for the visitor. Long paragraphs about company history and "our values" answer questions nobody's asking. Lead with what the visitor actually wants to know: what you do, whether you're right for them, and how to get in touch.

Skipping analytics. Without it, every decision about the site is a guess. It doesn't need to be complicated. It needs to exist, and someone needs to actually look at it occasionally.

Fabricating trust signals. Fake testimonials, invented client logos, or vague "trusted by hundreds of businesses" claims with nothing behind them tend to backfire. If you don't have real proof yet, say less and say it honestly.

Copying a template without adapting it. A generic hero-plus-features-plus-testimonials layout might be a fine starting structure, but if it doesn't reflect what your specific visitors actually need to know, it's decoration, not a system.

Your practical next steps

1. Write down, honestly, the three or four questions a real visitor has in mind before they land on your site. Check whether your current pages actually answer them.
2. Time how long your site takes to load on an ordinary phone on mobile data, not on office wifi. If it's slow, that's costing you visitors before they read a word.
3. Find your contact path right now, as a visitor would. Count the clicks from the home page to actually reaching you. If it's more than two, that's the first thing to fix.
4. Check whether analytics are even installed. If they're not, that's step one, before anything else on this list.
5. Look at your trust signals with fresh eyes. If anything on the page is exaggerated or unverifiable, cut it. If you have real proof you haven't used yet, add it.

None of these require a rebuild. Most of them are a focused afternoon of honest editing.

Frequently asked questions

Do I still need a website if most of my business already comes through WhatsApp or referrals?

Yes, usually. Referrals and WhatsApp are real channels, but both depend on someone already knowing the business exists, which puts a hard ceiling on how many new people can find you. A working website is what lets those same channels scale past word of mouth: it's the place a referral checks before calling, and the place a WhatsApp forward gets sent when someone wants the details.

How many pages does a small business website actually need?

Fewer than most owners assume. A handful of pages that answer real visitor questions, home, what you do, pricing or how it works, and a way to make contact, outperforms a dozen pages built to look thorough. Add a page when there's a real question it needs to answer, not to hit a round number.

Is a Facebook or Instagram page enough instead of a website?

They do different jobs. A social profile is useful for reach and updates, but the platform controls it, not you: the algorithm, the layout, and what's visible can all change without warning. A website is the one place the business fully controls, which matters most for Job 1 (answering questions plainly) and Job 4 (being found for specific searches), things a social profile alone can't reliably do.

How often should a business website be updated?

Small, regular updates beat a full redesign every few years. Prices, services, and hours should be corrected the moment they change, and it's worth checking the whole site every six months or so to confirm the contact path and trust signals still hold up. Treating launch as the finish line is one of the most common mistakes covered above.

What's the minimum a business website actually needs to work?

The five jobs above, not a page count. Plain answers to what you do and what it costs, honest proof you're real, a contact path that takes one or two clicks, enough structure to be found for the searches that matter, and analytics so you can tell whether any of it is actually working.

Where we can help

Business websites is one of our core services because this gap, between a site that exists and a site that works, is where most of the value actually lives. We build sites around the five jobs above from the start: fast on real connections, clear about what you do, honest about proof, easy to contact, and measurable once it's live.

If you want to see how we apply this thinking to our own site, not just other people's, we wrote about rebuilding roblemedialab.co.ke with the same tools we sell. And if the site's job is really to support a wider content and search effort, a content system, not just a calendar is the natural next read.

If you'd rather check it yourself first, the Lead Leak Checklist walks the six stages between a stranger on your site and a paying customer, with the specific things to test at each one. It costs nothing and it doesn't ask for your email address.

If your current website exists but isn't really doing anything, book a 30-minute discovery call and we'll look at it honestly: what it's getting right, what it's quietly costing you, and what's worth fixing first. No pressure, no hype.