Case Study
Massage practice website rebuild
A slow, English-only website was costing the client customers. I rebuilt it to be found, fast, and easy to run.
The client
A massage and sound-healing therapist in Ede came to me with a problem. Ten years in, more than 1,200 clients, five stars on Google. The people who found the practice loved it. The trouble was getting found in the first place.
The client had spent years trying to fix the website alone. Evenings and weekends went into wrestling with WordPress instead of doing the actual work. It was clearly holding the business back, and it would not budge.
The problem
The website looked fine on the surface. Underneath, it was costing the client customers three ways:
- It spoke the wrong language. The client is a Dutch therapist with Dutch clients, and the entire site was in English. So when someone nearby searched for a local massage, Google had to work to understand that this English site was their option. Every booking that starts with a Google search was starting at a disadvantage.
- It was painfully slow on a phone. On mobile, the main image took 12.6 seconds to appear. People give a website about three seconds before they give up. Most visitors were on their phones.
- It was full of dead pages. Google was crawling 68 pages, and a third of them were empty or duplicates left behind by the old WordPress setup. That kind of clutter makes a business look neglected to a search engine.
The rebuild
I rebuilt the site from the ground up so it would be found by the right people, load the moment someone taps it, and be something the client could run without a developer.
- Dutch first. The site now leads in Dutch, with a clean English version alongside it, and the behind-the-scenes signals that tell Google plainly: this is a Dutch business, for Dutch customers.
- Fast. It loads almost instantly on a phone now. The mobile speed score went from 58 to 97 out of 100, and that 12.6-second wait dropped to 2.6.
- Clean. I cleared out the dead and duplicate pages, rebuilt the real content properly, and added new pages worth having.
- Easy to run. The client can edit any of it now, in a few clicks, without a developer.
- Nothing lost in the move. Ten years of history carried across with a redirect on every old link, so the rankings the site had built up stayed put.
The highlights
A quiz that picks for them
The client offers a lot: deep tissue, Thai yoga, lomi lomi, shiatsu, sound healing, cupping. New clients often freeze at the choice and book nothing. So I built a short quiz that asks a few questions and points each person to the treatment that fits them.
It started as a small experiment I added on a whim. It went over so well that the client now points people to it everywhere.
600 five-star reviews, all in one place
Reviews are the client’s strongest asset: more than 600 five-star ratings, scattered across Google and Treatwell where they did nothing for the site itself. I built a compiler that pulls them all in and shows them off across the pages, so the proof works where people actually decide.
A blog built to rank
Search rewards a site that owns a whole topic. I researched what people genuinely search around massage, sound healing and recovery, then built the blog in clusters: groups of related posts that link to and reinforce each other, so the site builds authority topic by topic rather than chasing one keyword at a time. That foundation is what is starting to pull in the search growth and AI mentions further down.
The results
Two months in, more people are finding the practice, and the ones who do are staying longer.
PageSpeed Insights · Mobile
Before old WordPress site
After the rebuild
| What it means | Before | 2 months in¹ | 6 months | 12 months |
|---|---|---|---|---|
| Times it showed up in Google / month | ~22,300 | ~57,200 +156% | ||
| Visits from Google / month | ~230 | ~600 +161% | ||
| Visits from everywhere / month | ~460 | ~715 +56% | ||
| Time spent per visit | 20s | 46s +130% | ||
| Visitors who actually engage | 54% | 62% +8 pts | ||
| ”massage ede” visits / month² | ~28 | ~47 +69% | ||
| Mobile speed score (out of 100) | 58 | 97 +39 | ||
| Homepage load on a phone | 12.6s | 2.6s −79% |
¹ Search numbers compare 1 June to 6 July 2026 against 1 April to 6 May, two equal windows either side of launch, skipping the first weeks while Google recrawled the new site. Engagement numbers cover the first 8 weeks (5 May to 29 June). Everything is shown per month. The 6- and 12-month columns fill in as the site matures, which is when search results really compound.
² Still the first-8-weeks figure.
Good numbers for the first two months. Search compounds over time, so this is an early read with plenty of room to grow.
Showing up in AI answers
More people now ask ChatGPT or Google’s AI for a recommendation instead of scrolling through search results. The new content is already being quoted back by Perplexity, Copilot, Grok and Google’s AI overviews. It is early days, but the site is built for where search is heading.
Back to the work that matters
The site runs itself now. The hours that used to disappear into the website go back to clients, and the thing the client spent years fighting is simply handled. That is the part that matters most.
The booking side has its own story. I also replaced the old booking plugin with a custom tool, and built a customer database and newsletter system that never existed before. That one is written up separately.
A look at the site
Like what you see?
Let's build yours