West Sussex · one-person studio
I'm a solo IT contractor who builds apps on the side. This is how I got here, and what I've learned along the way.
I'm Martin Smith. I've worked in IT for over 30 years in various functions and sectors. I've been an employee and for the past 16 years a self employed contractor. I've done software and hardware support, I've built servers, I've designed databases, written apps, worked on data warehouses and done data migrations. I have a wide range of skills and abilities learned through years of self taught experience.
Before the birth of my second child in 2012, I thought I should get myself a good camera so that I could take better pictures of my growing family. That was the beginning of my photography hobby. Within a couple of years, I'd upgraded to better and better cameras and started a small collection of digital and film cameras. During this time I was active on Twitter (now X) where there was a great photography community. Through this community I got in to doing group photowalks where we would meet up in person and walk and take photos. In 2017, I organised my first photowalk using my blog to announce it and get registrations. While this worked okay, I wanted a better way. At this time I was learning another programming language (Python) and I was looking for a project that I could use it on. This is when I came up with the idea for photowalk.me.
I built photowalk.me over a couple of months, coding for hours every evening after work, using Flask and MySQL. I released photowalk.me in February 2018 and within a couple of months I had over 200 users and hosted about 20 photowalks all over the UK. It was a free platform for free photowalks. The only attempt to monetise it was to add Google Ads. I didn't like how the site looked with them and hardly any clicks were generated so I removed them. This is how it stayed for 7 years. I paid for hosting and the site ran itself, I maintained it and added the odd feature now and again. By 2025, it had grown to about 1,400 users and quietly ticked along.
At the beginning of 2025, I thought maybe I could build photowalk.me into something more and turn it into a real business and actually monetise it. So with the help of AI I redesigned the site, adding loads of new features and subscription tiers. I released the new photowalk.me in May 2025. The primary ethos has always been and always will be that free photowalks should be free to host. Whatever was free before would still be free, but if you want to charge for a photowalk, you need to pay.
The redesign and new features did initially bring in more users and more photowalks, but not much in the way of paying users. I was hoping to compete with the likes of Meetup.com and Eventbrite, but I overestimated demand. As of writing this (September 2026), there are 2,313 registered users with around 340 users active in the last month and £39 MRR. The issue has been finding organisers that want to charge, Instagram is hard to search, I've tried scraping Meetup.com, but data is often out of date. The few times I have managed to find people, I've messaged them and they have been really responsive and I've gained a few extra users. It's just not possible to do at scale. So for now photowalk.me remains a passion project I will keep it running for as long as people continue to use it, even if it never makes money like I had hoped.
In January 2024, a friend I knew through photography introduced me to someone looking for help building an app. He knew I'd built photowalk.me and thought I could help. The initial brief was a web app to import client data and generate charts in a specific format for client presentations. Within a day I had a proof of concept, then built a first version over a couple of months. It wasn't perfect, but it worked. From the start I told him front-end design wasn't my area. Photowalk.me was functional, not pretty, and my day job is mostly databases.
Shortly after delivery, the direction pivoted to a self-service B2C app. This was a different beast, needing a redesign, file ingestion pipelines, notifications, and generative AI. There was talk of raising funding, and outside design and engineering companies were brought in who wanted more money than was available. So I quoted again for an MVP that could later be replaced once funding came through. I did a lot of that work for free, chasing equity instead of pay. My first mistake was not getting that equity split agreed upfront. I eventually got a small percentage, with vague talk of a salaried CTO role if funding landed.
The MVP shipped to a small waiting list. It worked reasonably well but was buggy, and honestly I was out of my depth. What was being asked for was more complex than I could deliver solo, even with AI helping more than it does now. I kept fixing issues as they came up. Then a code audit, brought in by the founder, turned up a few security issues I'd missed. He lost confidence and stopped development. I was glad, in a way. The equity never felt like enough, and the funding never looked likely to arrive. I'd built what I promised, but he wanted something better, which was fair.
A year later, the final invoice was paid, I handed over the code, and we shook hands. He's since released a new app under different ownership. It looks considerably more polished than what I built.
What I took from it: I know my limits now, and I know I work better with control over what I'm producing. From here, I wanted to build what I decided to build, not what someone else changed their mind about.
As much as I enjoy the work I do in my day job, contracting is not what I want to do forever.
I've been looking for ways to supplement my income and maybe one day replace contracting entirely. I looked at offering more of the type of work I do as services to multiple clients instead of the one or two I currently do through contracting, so I started to put together a consulting offer.
I've realised, however, that's not what I want to do. Even with the issues around the health app, I enjoy building products rather than selling my time. My day job currently provides what I need to pay the bills, but it's not what I want to be doing long term.
I work out a lot, my health is very important to me. I turned 50 this year and I'm trying to get in the best shape I've ever been in.
Along with exercise and good diet, I take a number of supplements. Knowing what supplements to take can be difficult. So I decided to build an app that uses AI to make suggestions based on a simple questionnaire. With the help of Cursor, I built my first mobile app for Android and iOS, SupplementAI. I could have written all the code myself in React Native, but it would have taken months as I've not used it before. With AI, it took a few weeks. Even with AI, it's not as easy as you think. Having a coding background really helps. I initially released SupplementAI to the Play Store only, as I didn't have a Mac for an iOS build.
I strongly suspect I have ADHD, although I've never been diagnosed. I started using Claude to help keep me accountable for the tasks that I needed to do. I set up a scheduled task that would ask me what I did that day and what I planned to do the next day. That kind of worked, but if I didn't do what I said I would do, I would avoid answering out of shame. I wanted something that witnessed without judgment. I had recently purchased a MacBook Pro to do iOS builds of apps, so I created PactKeeper. You have one commitment per day and every evening you check in as kept, missed, or partial. As long as you answer one of these, you build a streak. The streak only breaks if you don't answer. My friend Darren, who has similar issues to me, thought it was a great idea and would test it for me.
Both apps are now in both the Apple Store and Google Play Store.
My contracting work is the floor. The apps I build are the goal.
The ultimate goal is to make enough income from a portfolio of apps that I can leave contracting. That's a long way off yet. For now, the target for January 2027 is small: 10 trial starts and 3 paying subscribers for each of the two mobile apps. That should be achievable. Where I stand in January will tell me what to do going forward.
Current numbers for these apps are public at /open.