Software doesn't end at launch: the part nobody budgets for
You think the big bill is building it. Then you learn that launch is the start, not the finish, and that keeping it alive has its own cost.
A lot of people treat launching their software as a finish line. It's finally done, it works, on to the next thing. That's understandable, because all the energy and all the budget went into getting there. The trouble is that with software, launch is more like the starting line.
Software that actually gets used never sits still. The world around it shifts under your feet. The phone ships a new version and something stops looking right. A law changes and you have to adjust how you store data. Another tool you connected to changes its rules. None of this is anyone's fault, it's what happens when you're alive and wired to other things.
Then there's what real users find. However well you tested something, the moment real people touch it, bugs and requests show up that nobody imagined. It's proof the software finally met reality, which is always stranger than any plan. It happens every time, and it's worth planning for.
The expensive mistake is treating maintenance as a surprise. The thing gets built squeezing the last euro of the budget, and when the first serious problem shows up there's no money and nobody assigned to deal with it. So it sits untouched, slowly rots, and a year later it costs more to fix than it would have cost to look after from the start.
A rough rule that works: set aside, for the first year of its life, a slice of what it cost to build, meant only to maintain it. That money has one job, to keep what you already have healthy, not to add things. It sounds like an expense, but it's what stops your investment from rotting right when it starts to pay off.
When someone asks us for a quote, we try to talk about this from the start, even though it isn't the part the client wants to hear. Software is more like a plant than a piece of furniture. You buy it once, yes, but if you don't water it, it dries out, and nothing is more expensive than rebuilding something you let die to save yourself the watering.