How a $2,400 redesign paid for itself in 11 days

The site had been live for six years. Diane's nephew built it when she opened the bakery. He did his best. But six years later it was loading in 8.4 seconds, the phone number was an image you couldn't tap, and the menu was a PDF from 2019 still listing items she'd stopped making two seasons ago.

Diane wasn't sure she needed a new site. She thought the problem was that not enough people knew about her. I looked at her Google Search Console data for about ten minutes and told her the opposite: people were finding her just fine. They were just leaving.

What the old site was actually doing to her

Before touching anything I ran the same audit I do for every client. A few things stood out fast.

The homepage had a full-width photo of the storefront taken at dusk. Moody, sure. It was also a 4.1 MB JPEG with zero compression. That one image accounted for most of the load time. On a phone with a decent LTE signal it took six seconds just to paint the picture. By then, 61% of mobile visitors had already bailed, according to her analytics. They weren't going somewhere else to find a bakery. They were just gone.

The contact page had a form that emailed a Yahoo address Diane hadn't opened in two years. Custom cake orders were going into a void. She knew she was missing some, but not how many. A rough count from her saved messages suggested at least 34 custom cake inquiries in the previous year had vanished without a reply.

And the menu. The PDF wasn't searchable, didn't work on mobile, and featured a "seasonal special" from October 2019. Customers kept showing up asking for things she no longer made. Annoying for everyone, including the person behind the counter who had to explain it.

What actually changed

I rebuilt the site from scratch. New design, new structure. I'll spare you the technical walk-through because it genuinely doesn't matter to you. What matters is what changed for Diane's customers.

The homepage now loads in 1.1 seconds. The dusk photo is gone. In its place: a tighter shot of the actual pastries, compressed to 180 KB. The phone number is a real tap-to-call link. That sounds obvious. It wasn't there before.

The menu is a real page now. Diane can update it herself in about three minutes using a simple tool. No more PDFs. No more ghost menu items haunting the counter staff every Saturday morning.

The contact form goes to an address she actually monitors. I also wired up a simple auto-reply so customers know their order request landed. That matters more than it sounds. It kills the "did anyone get this?" anxiety that makes people call instead, which eats up more of Diane's time than the form ever did.

If you want to understand exactly why slow load times cost real money, this piece on what a slow site does to your traffic lays it out plainly. Short version: people don't wait. Not even a little.

The numbers, three months later

Diane and I checked in at the 90-day mark.

Mobile bounce rate dropped from 61% to 23%. More than twice as many people were now sticking around long enough to look at the menu or reach out. Custom cake inquiries through the form went from near-zero (because the old form was silently broken) to 17 in the first month, then 22 the next.

She tied $6,200 in new custom orders directly to the site over those 90 days. Some of that is seasonal, and she said so honestly. But she'd never had a custom order stretch like that before. Her average cake runs about $145, which works out to roughly 43 orders she wouldn't have seen otherwise.

The site cost her $2,400. It paid for itself inside the first 11 days of custom orders.

The thing that surprised her most

Diane told me the biggest change wasn't the revenue. It was the phone calls. She used to get a steady stream of calls asking "do you still do almond croissants?" or "what are your holiday hours?" After the redesign those calls dropped sharply because the answers were right there on the site, easy to find, on a phone.

Each of those calls cost her 3 to 5 minutes. She was fielding maybe 8 of them a day. That's up to 40 minutes a day answering questions the website should have been handling all along. Over 90 days she got back somewhere around 60 hours of her life.

That's not a rounding error for a person running a bakery mostly by herself.

What this means if you're in the same spot

A lot of small business owners I talk to have a version of Diane's situation. The site exists. It technically works. But it was built a while ago, probably by someone doing you a favor, and it hasn't kept up. Meanwhile your competitors have sites that load fast on phones, have working contact forms, and show up better in local search.

The question isn't whether to fix it. The question is how long you've been paying the invisible cost of not fixing it. Missed inquiries, phone calls eating your afternoon, customers who looked you up and picked someone else. That cost is real. It's just spread thin and hard to see all at once. The real cost of a good-enough website goes deeper on this if you want the full picture.

It also helps to know what's specifically turning people away before you start. Here are some of the most common signs a site is costing you business, most of which owners miss entirely because they're not the ones using it on a phone for the first time.

And if any of this sounds familiar, tell me what you have and I'll take a look. No pitch, just a conversation.