How do you prioritize software stability over new feature requests without stifling innovation?

We all have 'values'. The values differ in priority depending on who the audience is. How each value 'stability' or 'innovative' is measured, prioritized etc depends on audience appetite, product maturity, engagement factors (methods channels) and how well communication is received through whose approaches. Too many times I have seen developer processes (born out of convenience) delivered into customer systems, that just don't match - what is good for one, is not good for another...

Align your software priorities and deliver with the prioritized agenda of your overall software/product vision. Pick enhancements or new features that aligns with it thereby leading to implicit buy-in from executive leadership and the business.

Based on my experience, using agile methodologies allows you to advance in developing innovative features while maintaining the stability of your system's core. Identify the MVP that meets user expectations, engage with them, assist them in simplifying their processes and detect unnecessary functionality. Once you implement it, it is very likely that no additional iterations will be requested as it meets its goal, simplifies their activity, and simultaneously allows you to gradually incorporate innovation, with a noticeable system growth in short periods of time. When we first started using Scrum, three years ago, we planned developments in four or five iterations; nowadays, we know that our users will be satisfied with just one or two.

Is development really necessary, or is automation sufficient? This is the first question we ask ourselves after understanding our users' needs. If you can deliver the same value with simple automation in a short period of time, do it. Once they have it, other needs may arise that will make you reconsider development but with a much clearer vision of what they need. In other cases, you may simply eliminate the need with this simple operation, all the options lead to positive outcomes. For me, the most difficult part of development is identifying what exactly the user needs, even when they themselves are not aware of it.

Sometimes, even if the team fosters a culture based on quality, new members may initially overlook this principle. No problem, they will soon experience the issues arising from these bad practices and will join the group. If not, it can be beneficial to assign them poor code to work on or present them a challenging situation so they can see the impact.

Take small but frequent steps with limited visibility to key agreeable users at each stage of the CI/CD pipeline. Consider also the impact of what multiple upgrades means; backward comparability, visibility and impact not just to the technical features but hidden / unintended features users may actually like. Continuous feature monitoring can provide insight and take-up.