Behind the Scenes
Build decisions, design lessons, and real advice for small business owners who want to grow online. No corporate language. Written by someone who does the work.
Most small business websites fail silently. No error messages, no obvious broken links — just a slow, forgettable experience that sends potential customers straight back to Google. Here's what's actually happening.
Five projects. Five completely different clients. Five sets of things that went wrong and had to be fixed. Here's what those builds actually taught me about design, clients, and what makes a website work in the real world.
Most small business websites fail silently. There's no error message, no obvious broken link, no moment where something visibly goes wrong. The site just sits there — slow, forgettable, and quietly sending potential customers somewhere else.
The business owner usually doesn't know this is happening. They're busy running the actual business. The website got built once, it went live, and it's been sitting there ever since. Meanwhile, every person who lands on it and leaves without calling is a customer they'll never know they lost.
"Your website isn't just a digital business card. It's working — or failing — for you every single hour of every day."
Studies consistently show that more than half of mobile users will leave a page that takes longer than three seconds to load. Most small business websites — especially those built on bloated page builders or old WordPress themes — take five, six, sometimes ten seconds to fully load on a mobile connection.
That's not just an inconvenience. It's a conversion killer. The visitor came from a Google search. They had a question. They landed on your site, waited, and went back to try the next result. You never even got a chance to speak to them.
A website without a clear call to action is like a shop with no cash register. People come in, look around, and leave because there was no obvious path to buying.
Most local business websites suffer from this. There's a homepage with some information about the business, a services page with a list, and a contact page buried in the navigation. No urgency. No direction. No reason to act now rather than later — which almost always means never.
If your website looks like it came from a template — and your visitor can tell, even if they couldn't explain how — you've already lost a significant portion of the trust battle. Customers make snap judgments about businesses based on their website in seconds. A generic site signals a generic business, regardless of how good the actual service is.
"A website that looks like everyone else's tells the customer there's no reason to choose you over everyone else."
It loads fast — under two seconds on a mobile connection. It has one clear primary action above the fold that tells the visitor exactly what to do. It looks like it was built for this specific business, not pulled from a library of pre-made layouts. And it continues to work at every scroll depth — not just the hero section.
Go to your website right now on your phone, on a normal mobile connection, not your home WiFi. Time how long it takes to load. Then ask yourself: if you were a stranger looking for what you offer, would you wait? Would you trust what you saw? Would you know what to do next?
If the answer to any of those is no — the site is costing you. Not in a dramatic, visible way. Quietly, every day, one lost visitor at a time.
The good news is that this is fixable. And it doesn't require spending big-agency money to fix it.
Share your current site and we'll give you an honest assessment — what's working, what's hurting you, and what the most effective fix would be.
Five projects. Five completely different clients. Five sets of assumptions that turned out to be wrong, problems that had to be solved on the fly, and lessons I couldn't have learned any other way.
This is what actually happened — not a polished retrospective, but an honest account of what building real websites for real clients taught me about design, about communication, and about what makes a website work in the world rather than just in Figma.
Every client came in with a brief. "I need a website for my cleaning business." "I want a portfolio." "We're a startup, we need to look credible." Clear enough on the surface. But in every single case, the real brief only emerged through conversation.
The cleaning business owner didn't just want a website — she wanted to stop losing jobs to a competitor who had a slicker online presence. The portfolio wasn't just a showcase — it was an attempt to transition from freelance work to agency clients. The startup founder needed the site to work as a pitch deck substitute for investor meetings.
"The stated brief is where the conversation starts. The real brief is what you find when you ask the right questions."
Now I spend the first conversation almost entirely on questions. What does success look like six months after launch? Who exactly is the person arriving at this site and what are they trying to decide? What happens if this site doesn't work?
I knew this intellectually before I started. I knew the statistics. I still caught myself designing for desktop first on two of the first five projects. The result both times was a mobile experience that felt like a compressed version of something else rather than something designed for the screen it was being viewed on.
The fix wasn't complicated — start every layout decision from the smallest screen and expand outward. But it required a genuine shift in how I was thinking about the work. The mobile version isn't the adapted version. It's the original.
On the third project I built something I was genuinely proud of visually. The animations were smooth, the layout was distinctive, the typography was doing exactly what I wanted it to do. Then I ran a Lighthouse audit and saw a performance score of 58.
The culprit was unoptimised images and three fonts being loaded when one would have served the same purpose. Fixing it took about four hours. The score went to 91. The site looked identical. The lesson stuck permanently:
I learned very quickly that describing a layout in words is almost completely useless. "The hero will be full viewport with the headline on the left and a visual on the right, dark background, purple accent." The client nods. They have no idea what this will look like. Then they see it and ask why it's dark.
Now I show references before I start. Three to five examples of sites with the visual direction I'm proposing. Agreement on references means agreement on direction, which means far fewer surprises when the first version lands.
The fifth project taught me this. Great site. Client loved it. I handed it over with a written document explaining how to update content. Three weeks later I got a message — they'd accidentally broken the navigation and had no idea what they'd done or how to fix it.
Now every project ends with a 1-hour call where I walk through the site live — what each section does, how to update text and images, what to do if something looks wrong, and who to contact. The handover isn't an admin step. It's the moment the client goes from having a finished site to actually owning it.
"A website the client can't confidently use is only half a delivered project."
Five projects in, the thing I'm most certain of is that the work gets better every time — not just technically, but in terms of understanding what clients actually need versus what they say they need, and how to design for outcomes rather than aesthetics alone.
The sites that have performed best weren't necessarily the most visually ambitious ones. They were the ones where the design decisions were most directly connected to a specific business goal. That's the standard I'm building every project to now.
The portfolio has live links to every project — click through and see how they perform in the real world.