Don't Try This At Home

Especially in small companies there is often a lot of reward for "saving the day" and "putting out fires", above and beyond the sheer thrill of it. Fixing a disaster is a lot more visible than preventing one, and it generally reduces the consequences of poor planning or bad engineering further back along the funnel. Your team "putting in the extra effort" is also visible, so it usually ends up being rewarding at the level of management that is supposed to be defending against short-term thinking.

Firefighting also usually involves aggressive shortcuts. It's worth looking at these as indicators of things you can improve in your normal processes - otherwise it can support the feeling that firefighting time is "better" than normal development time. In the least-bad version of this, putting the fire feels productive but only defers the actually-needed maintenance work for a little while.

Each bit of firefighting should have a high priority post-mortem analysis, which should cover the problem itself and the shortcuts used by the solution. Both kinds of knowledge should lead to bug reports and actions to be taken - you want to do a better job of not having that kind of problem again, but you also want better support for off-the-wall problems in general. Do you need better logging and diagnostics? More "observability"? Does it need to be easier to test a one-line change to your entire system? Perhaps you need a more accessible list of on-call specialists, divided more clearly into subsystems - or if you have such a list, maybe the organization needs to better reflect the division of components from an on-call perspective, instead of an architect's perspective. What you don't want to end up doing is codifying the firefighting as the proper way to respond to this particular problem.

Firefighting isn't entirely bad - if you have a customer-facing product, being responsive (and being seen to be responsive) is good, though not as good as having higher quality earlier in the first place. The real problem is that it's a bad habit to develop, and there are systemic and cultural feedback loops that encourage the habit - you need to continue to apply pressure to avoid getting caught by the downsides.