The right solution isn't always the best solution
The thing about iterating towards a long term solution via a series of minimum viable changes is that you have to deliberately start with a solution that you know is wrong. Maybe that means starting with an excel template instead of a LIMS. Maybe it means having your ML team manually run the inference step instead of putting it behind an API. Maybe it means hacking together UI with a low-code framework instead of building a custom React app.
In all these cases, you know what the right solution is. But the best solution for each one has an important characteristic: You can get it working and start iterating in days rather than weeks or months. Getting to a quick solution builds trust with your stakeholders and gets them on board with the approach. Iterating quickly allows you to both nudge the solution towards your stakeholders’ mental models and nudge their mental models towards the right solution.
Bad software is a symptom, not a cause. If your organization isn’t ready for the “right” solution, then maybe it isn’t the right solution.
It takes courage to trust that you’ll be able to iterate to a solution that you can be happy with. It takes patience to slowly nudge your organization in the right direction. But ultimately, that’s the only way you’ll get everyone to where they need to be.