I used to subscribe to a lot of software. Clipboard managers, remote desktop tools, update checkers, financial trackers, health dashboards, CRMs. Over time the subscriptions piled up and the monthly cost became genuinely annoying.
But cost was only part of the problem. The bigger issue was that none of these tools did exactly what I needed. They all did too much, or not enough, or the wrong things in the wrong order. And every few months they would push an update that changed the interface and I would have to relearn how to use something I was already paying for.
So I started building my own.
The tools I replaced
I like to keep track of things I copy throughout the day. I tried Paste, Awesome Clipboard, and a few others. They were bloated, complicated, had too many options, and the updates kept changing how they worked. I built Sniptique - a simple clipboard manager that does exactly what I need and nothing else.
I was using Chrome Remote Desktop and Jump Desktop to access my Mac when I was away from home. They worked, mostly, but Jump Desktop is a subscription and Chrome Remote Desktop has always made me nervous about losing the connection when I am travelling with no way to fix it. I built Telepathique - my own remote desktop app that handles screen sharing, keyboard and mouse input, and file transfers. No subscription. No features I do not use. And if something goes wrong with the connection, the app can handle it intelligently rather than just dropping me.
I used to use MacUpdater to keep my software current, but it got deprecated. The replacements were either expensive or overcomplicated. I built FreshenApp - a small menu bar app that monitors my installed software for updates and notifies me. That is all it does.
I find interesting links constantly while browsing - articles, tools, resources - and I never had a good place to put them. I did not want a full-blown app like Notion or Things. I just wanted something that would capture links as I copied them, summarise what each page was about, and let me get back to them later. I built LinkMinder, which does exactly that and emails me a daily digest.
I have always struggled to keep track of my finances and my health in one place. The commercial tools like MyFitnessPal or various banking dashboards are expensive, complicated, and never show me exactly what I want. I built a personal dashboard that pulls my bank balances daily, monitors my subscriptions, predicts my future financial state, and lets me query it. For health, I built a separate dashboard that pulls data from Apple Health - blood sugar, weight, steps, blood pressure - and graphs everything the way I want to see it. I was type 2 diabetic. I needed specific information presented in a specific way. No commercial tool gave me that.
I also have about 20 to 30 years of diary entries. I built a tool that treats my entire diary history as a searchable knowledge base, so I can look for patterns, retrieve specific entries, and learn things about myself that I had forgotten.
And most recently, I built Synaptique - a local knowledge base that sits on my Mac and monitors what I am working on. When it recognises that something I am doing relates to a link or resource I saved weeks ago, it surfaces it. It is basically a memory extension. It runs a machine learning model locally, so nothing leaves my computer. No commercial tool does this in the configuration I wanted, so I built it from scratch.
Why I stopped subscribing
The breaking point for me is usually one of three things.
The cost. I am paying a monthly fee for an app where I use maybe 20% of its features.
The bugs. The app breaks, does not work properly on my machine, or just does not do what I need it to do.
The breaking changes. The developer pushes an update that completely changes the interface, plugs AI into everything whether I asked for it or not, or removes a feature I relied on. The app I was paying for is no longer the app I signed up for.
When any of those things happen, I start thinking about building my own. And with the tools available now, a working first version usually takes me a day or two. Another couple of days to refine the design and the interface. Within a week I have something usable that does exactly what I need.
This is not just a nerd thing
I build these tools for myself because I can. But the same logic applies to any small business.
Think about a restaurant paying for booking software. Those tools are expensive, complicated, and full of features the restaurant does not need. It would be straightforward to build a custom booking system for that restaurant - plugged into the payment systems they already use, branded to match their business, fast and simple to use - for a one-off cost.
The same applies to hairdressers, tradespeople, estate agents. Anyone using a CRM like Monday.com or HubSpot is paying for a complicated, bloated tool that forces you to pay per user. Some businesses only need a fraction of those features. They still need the tool to interact with their other services, but they only need a minimal set of CRM capabilities.
If you are paying 50 to 100 pounds per month per seat for a premium tool and you have 5 to 10 employees, that could be 1,000 pounds a month. I can develop a custom tool for that price as a one-off cost. Bug fixes are free. Feature additions are done on a fixed, predictable cost. You get a tool that is exactly suited to what you need, at a price you understand, and it scales without charging you more to add more users.
The objections I hear
When I suggest this to people, I usually hear three things.
“I don’t trust AI.” Fair enough. But AI is just one of the tools I use to build faster. The end product is real software, tested and documented, running on standard infrastructure. You do not have to trust AI. You just have to trust the developer (I was there when the internet started…), the finished product, and I can show you that working before you pay for it.
“I’m used to the existing tools.” That is the most common one. Familiarity is powerful. But familiarity does not mean the tool is good. It means you have learned to work around its limitations. That is not the same thing.
“I didn’t know software could be built this quickly.” This is the one that surprises people most. The development landscape has changed dramatically in the last few years. What used to take months and cost tens of thousands now takes days or weeks and costs a fraction of that. Most business owners have not caught up with what is actually possible.
Where custom builds do not make sense
There are situations where bespoke software is not the right call. If your industry requires certified tools with specific regulatory compliance - medical records systems, financial trading platforms - you should probably stick with the established software. If a tool relies on a proprietary file format or a locked API that you cannot replicate, a custom build might not be viable.
But in most cases, the tools that small businesses pay for every month could be replaced with something custom that costs less, does more of what they actually need, and does not change underneath them every time the vendor pushes an update.
The other thing businesses should think about is reliability. I provide full documentation on every tool I build, written to industry standards, so that if something happens where I am not available to maintain or update the tool, it can be handed off to another developer without any problems. That is an important part of doing this responsibly.
If this sounds familiar
If you are a business owner reading this and thinking “I pay too much for software that does not do what I need” - get in touch. Tell me what tool you are using. Tell me what you wish it could do, or what it does that you wish it would stop doing, and how much it is costing you. Let me come back to you with a suggestion for something custom that is going to be cheaper, more effective, and built around how your business actually works.