From a problem I had to 500 users in seven months

I'm Dominik. I run FluoSite, a small web design studio in the Netherlands, and seven months ago I built FluoTest to fix a problem I had with my own sales calls. FluoTest is a free tool for building scored quizzes: people answer a few questions, get a score and a result, and you decide what happens next.

It just crossed 500 signups. I still build it alone, next to a full-time job, with no funding and no co-founder. This is what happened so far, what I shipped, how people found it and what I learned on the way.

Where it started

In February I was lying in bed, thinking about a problem I had with FluoSite.

I kept getting on calls with leads who were never going to be a fit. Wrong budget, wrong timing, wrong expectations. Every one of those calls cost me time I didn't have, because I was building FluoSite next to a full-time job.

I wanted to know who was worth a call before I got on one. The usual answer is a lead qualification quiz: a few questions about budget, timeline and needs, each answer worth points, and a total score that tells you how good a fit someone is.

So I built one. Just for myself. At that point that's all FluoTest was: a quiz, some scoring, and a way to figure out whether a lead was worth talking to.

Other people had the same problem

While I looked at how other people solved this, I kept running into the same posts. People looking for an alternative to ScoreApp, the best-known tool for this, because it cost more than they wanted to pay just to sort their leads. People hacking scoring together in forms and spreadsheets. People asking for something simpler.

I knew exactly what I would want it to look like, so I cleaned it up and posted it on Reddit.

The replies were things like "I've actually been looking for a ScoreApp alternative." That was the first real sign this wasn't only my problem.

User number 26

About a month in I had around 25 users and had more or less taken a break from it. I was in Spain with my girlfriend, not thinking about FluoTest at all.

Then a notification: new signup, user number 26. I was curious how they found it, so I opened Microsoft Clarity, the free tool I use to see where visitors come from and what they do on the site.

They came from ChatGPT. They landed on my FluoTest vs ScoreApp comparison, spent almost 19 minutes looking around, and signed up.

I hadn't messaged this person or sent them a link. They had a problem, asked an AI about it, and FluoTest was the answer they got. It was the first time I thought this might grow without me pushing every single user through the door. I wrote the full story in How ChatGPT became our biggest acquisition channel.

What I shipped

The product today does a lot more than the quiz I built for myself. Along the way the direction became clear: score any decision, automate what happens next. The quiz shouldn't be the end. If someone is a hot lead, something should happen. The main things I built:

Scoring and results. A quiz can score several things at once, sort people into tiers and show each person a result based on what they actually answered. A score map in the editor shows every score a quiz can reach before you publish it.

Explaining every result. Each response shows why that person got their result: every score, every tier and every rule that was checked. When a client asks why a lead was routed a certain way, the answer is right there.

What happens next. Results can send people to a booking page with their score filled in, deliver a file or a coupon, email them a PDF of their results, and push the lead into Google Sheets, HubSpot, Slack, Notion, Airtable or any tool through a webhook. Hot leads can trigger an alert the moment they come in.

Workspaces for agencies. Agencies and consultants can keep each client in a separate workspace, with the client's own name, logo and colours.

The full list is in the changelog. Most of it exists because someone asked for it.

How people found it

I have never paid for an ad. Every new account answers one question at signup: how did you find us?

AI assistants are the biggest answer by far. 205 of the first 499 signups said ChatGPT or another assistant sent them. The rest came from Google, social media and people telling their friends.

That didn't stop at user 26. I kept writing clear, honest posts that answer the questions people actually ask: comparisons with other tools, how-to guides and use cases. Most of my marketing is SEO for Google and GEO, which means making sure AI assistants understand what FluoTest is and recommend it. I wrote up what worked in How we got cited by ChatGPT and Perplexity.

It is starting to compound. A month ago FluoTest had about 370 accounts. In the 30 days since, 128 more people signed up, 19% more than the month before.

I still use it myself

The funny part is that FluoTest still solves my original problem.

New FluoSite clients fill out an onboarding quiz, get their results as a PDF by email, and I get everything I need on my side before we even talk.

It started as something I built for myself. Now I'm building the bigger version because other people want the same thing.

What I've learned

Your own problem is a good start, not the whole market. I built FluoTest for lead qualification. Then a vet in the UK used it to screen dogs before surgery, a safety officer in South Africa built compliance checks and a physiotherapist in Poland built a shoulder injury assessment. Scoring answers to make a decision is a much bigger category than the one I started in.

Clear writing is distribution. Most of my signups come from AI assistants recommending FluoTest. There was no trick behind it, just posts that answer one question plainly and honestly, including where other tools are the better choice.

Signups are the easy part. 103 people signed up more than three days ago and never built a quiz, and only about 6 in 10 people who build one go on to publish it. Getting people to their first real result is what I'm working on most right now.

You can do this next to a job. No big audience, no funding, no co-founder. Evenings, weekends, talking to users and shipping small things often. It's slower, but it works.

What's next

The goal is to take the whole process from the first question to the next real human conversation. Today that means a booked call or an email. Eventually I want it to include SMS and phone calls too.

Closer to home, I want more of the people who sign up to get to a published quiz and their first answers, because that's when FluoTest starts being useful.

Seven months ago I was lying in bed wondering whether this was even worth building. I don't know exactly where it goes from here. But I built something because I needed it, other people started needing it too, and it's slowly becoming a real product. That's the part I'm most proud of.

If you're one of the first 500: thank you. Every reply, bug report and feature request shaped what FluoTest is today.