Thoughts before they become conclusions.

I’m starting to think about software differently.

Building it is beginning to feel more like editing a website. You see something that could work better, change it, try it, and keep going. The distance between having an idea and putting something usable in someone’s hands is getting shorter.

For a long time, the cost of custom development pushed us toward fixed products. We chose a platform, learned its rules, and built our work around what it allowed. Changing the software was often a project of its own.

That relationship is starting to shift.

A case manager should be able to show us where a process wastes time. We should be able to respond while the problem is still fresh. A useful change shouldn’t always have to wait for a roadmap, a budget cycle, and three meetings.

But this also changes what makes a software company valuable.

If another team can recreate the interface quickly, having the nicest screens becomes a thinner advantage. I’m more interested in what sits underneath: how information moves, what connects to what, who can trust it, and whether the system actually helps someone take the next step.

Speed doesn’t settle those questions. Someone still has to understand the work. Someone has to test the changes, protect the information, and take responsibility when something goes wrong.

That’s where I want to spend more of my attention.

The opportunity is to build software that can keep learning from the people using it, without making them start over every time their needs change.

The tools will keep getting easier to build. Understanding what deserves to be built will still take care.