Perfectly good software

I’ve started making a lot of software, a lot of little tools, for myself, for my team and for my company. I’ve built tools for grooming our backlog, I’ve built tools for working on my mindset and I’ve built micro-tools just because they bring me joy.

For some reason, I keep describing them as “perfectly good”, and that phrase has really stuck in my mind. What does “perfectly good” mean in this context? I asked gpt-5 the meaning of the phrase, and it spat out the below.

“It’s often used defensively, to push back against unnecessary replacement, waste, or perfectionism”.

I like that!

So much of the software we pay for, whether for ourselves or our teams, is unnecessary, expensive, and often over-engineered. Internal tools and apps that glue together internal workflows, don’t actually need to be perfect, they just need to work.

The reality is, a lot of the software we use, isn’t actually that important. At least, it’s not so important that we should spend tens of thousands of dollars on it, and certainly not so important that we can’t manage to build and maintain our own version of it, now that the cost and effort required to do so has shrunk so drastically.

To be clear, you should definitely still buy a CRM system, an ERP system, and a payroll system. It makes sense to invest in deep and robust solutions for system-of-record-style products. But the average SaaS company now has something like 50+ apps in-use internally now, and there’s no doubt in my mind that many of those could be now bought in-house, streamlined, merged, and ultimately swapped out with “perfectly good” internal replacements, built for a fraction of the cost, and now fairly maintainable with AI coding agents and developers working in tandem.