These projects quickly reach a point where evolving it further is too costly and risky. To the point that the org owning it will choose to stop development to do a re-implementation which, despite being a very costly and risky endeavor, ends up being a the better choice.
It's easy to say that organizations should do it right the first time, in terms of applying proper engineering practices. But they often didn't have the time, capital, and skillset to do that. Not ideal, but that's often how things work in the real world and it will never change.
Organizations should do it not catastrophically wrongly, especially once a core design / concept is mostly solidified. Putting a little time into reliability and guardrails prevents a huge amount of downside.
I've been at organizations that don't think engineers should write tests because it takes too much time and slows them down...
This is a very costly way of developing software.