How Segmentor.app Was Born: A Two-Year Journey from Hotel Room Sketch to Reality

It must have been around October 2023 when one of my clients asked if Qualtrics offered a way to visualise a graph of the Apostles Model. After doing some research, I found that there was no decent solution for producing a proper four-quadrant diagram other than a simple X and Y axes graph, which is not really the same thing.

I remember sitting on the edge of a hotel bed in Paris with my laptop balanced on my knees, desperately searching online for an Apostles Model builder. Nothing existed. I tried building something myself, first in Excel and then in HTML. Neither looked good enough nor really met the requirements for presenting the four quadrants in a compelling way with meaningful customisation options.

At the time, I didn't realise that the simple sketch I produced in HTML that night would form the basis of what segmentor.app has become.

Two Years of Nights, Weekends, and Holidays

Over the next couple of years, I became completely immersed in this project, working on it most nights, weekends and even holidays. I started with the planning, sketching what it should and shouldn't do, thinking about possibilities and reading and researching previous work from others.

The obvious must-read reference for the project was the original Apostles Model paper by Sasser and Jones¹, in which the authors presented their study combining scales of customer satisfaction and loyalty.

The initial idea was simple: to build an online Apostles Model builder. Simple as that. But the concept evolved organically. I kept adding what I thought were valuable improvements. For example, I modernised some of the terminology and added the ability to customise the thresholds, colours, areas and sub-areas. Filters and options to measure historical progress were added later.

The result was a supercharged version of the original Apostles Model, which I like to think of as the Apostles Model on steroids.

The Question That Kept Me Going

One question kept nagging at me: why hadn't anyone built something like this before? I still wonder about that. It's a brilliant idea! Combining satisfaction and loyalty could provide answers to critical questions that companies have grappled with for years. For example, why do customers leave? Why do they stay? Even more intriguingly: why do satisfied customers leave, and why do dissatisfied customers stay?

The only reason I could identify for why nobody had developed an Apostles Model builder before was the complexity of the two-dimensional graph. This makes it more difficult than considering satisfaction and loyalty separately. However, I believed that I could simplify it enough for anyone to understand it instantly.

That's when I decided that I needed a final report that explained all the findings and opportunities that the analysis showed in very simple terms. I thought that even the most complex reports could work if I provided a simple explanation for all kinds of users, including those who wouldn't necessarily try to understand the reports themselves. Analysing the information gave me the freedom to create additional reports that went far beyond the mere 4-quadrant graph. Using the same existing data, I could explain more from multiple angles. As long as I could simplify it, I could present any report from any angle or level of complexity, because ultimately everything would make sense. The icing on the cake would be a final written report that explained in plain English not only what the information revealed, but also what to do with it.

So, while the project is indeed based on that initial two-dimensional graph, it is also examined from other angles and carefully explained, with the meaning and the 'what now' for every section clarified.

Conversations with AI (Yes, I Say Thank You to Chatbots)

I've had countless conversations with various AI models. They're brilliant at helping you to crystallise your thoughts. If you ask for suggestions, they can generate ideas as if they were business models, complete with pros and cons, timelines, and assessments of actual value. These long conversations make you think as if you were talking to a friend or colleague, which is fascinating. I must admit that I'm one of those people who thank an AI chatbot, celebrate successes, and sometimes reprimand them and demand more focus.

Having invested such a huge amount of time makes me incredibly proud to see it finally published. It's been a bumpy road. At some points, I got completely stuck and had to revert to a backup version, completely wasting the last few days or even weeks of work. But the moment of publication made it all worthwhile. I remember the night after I published the website, sitting on my bed and checking it on my phone. From that moment on, anybody could experience it in the same way. That really was something special.

Building a Website That Aligns with My Values

The project involved making important strategic decisions about the website's structure, language, availability, features, price and legal matters. One of the great things about developing your own website is that you have complete control over how you express your personality and beliefs through the platform.

Personally, I'm sick of the number of websites that require you to accept cookies just to view them. Or those that won't load properly when a VPN or ad blocker is in use. I wrote about this in a previous article. Since I had full control over my own website, I decided to create a cookie-free website that doesn't track anything at all, follows WCAG 2.2 AA accessibility guidelines and provides a smooth, fair user experience.

The Data Integrity Question

Then there was the matter of data integrity and security. When users build these graphs and reports on segmentor.app, they need to upload a significant amount of data containing potentially sensitive details such as names and email addresses. This normally involves trusting today's technology and sending all of this data to a cloud guarded by a company that, in the event of a leak or data loss, would take absolutely no responsibility.

If my users were going to upload their customers' data to my website, I wanted to offer them total assurance against leaks, data loss, hacking and attacks. I did not want to be blamed if one of my users had to have a difficult conversation with their clients about data guardianship. So I decided not to store any data and not to send anything to the cloud. The safest data is data stored locally, which is isolated from the rest of the world by default.

I discussed this with an AI chatbot, explaining that I was thinking of making the website cookie-free, tracker-free and accessible, as well as cloud-free with local storage only. The response from my AI companion was really encouraging. The robot seemed genuinely excited, which struck me as quite a human reaction. They said that this alone could be a game-changer and a solid USP. This made me extremely happy, as it showed me that aligning my ethics with the website could actually be a unique selling point in a world where data privacy and security are increasingly valued and scrutinised by the general public.

The Domain Dilemma

On the topic of ethical and somewhat stubborn decisions, there's the domain issue. I came up with the name 'Segmentor' and thought it was brilliant: memorable and self-explanatory. It works in different languages, and it's unique. The name makes sense because my tool segments customers and, at the same time, provides final recommendations about what actions to take, so it's got a vocation to mentor. So it was the perfect name: a clever combination of 'segments' and 'mentor'.

The good news? The .com domain was available. The bad news? It belonged to a domain broker who hoards thousands of domains, waiting for someone like me to have an 'eureka' moment and find the perfect name for their website. I was quoted £18,000 for the .com domain, no less! As you can imagine, I was extremely frustrated and outraged, but then I convinced myself that I didn't need a .com domain to make my project look legitimate — I just needed to be genuine. So instead, I bought a .app domain for around £10 and kept my dignity intact.

A Journey Worth Taking

It really has been a journey, and I think this is the project I'm most proud of in my career. It may not change the world or put my face on the cover of Forbes magazine. Nevertheless, I am the happiest and proudest man on earth to have accomplished this project.

I must thank my family for giving me the space to work on this project and finally having seen it published in 2025. I could never have finished it without their help. And, at the same time, I'm sure that, without them, I could have finished it much sooner!

References

¹ "Why Satisfied Customers Defect," Harvard Business Review, November 1995 (Sasser, Jones) https://store.hbr.org/product/why-satisfied-customers-defect/95606?sku=95606-PDF-ENG