Visit dont.build.
Technical decisions are often about being as less-wrong as possible than being right. Decision parameters themselves are dependent on even more parameters, such as the business environment, expertise, team and user culture, cost ... There is no one-size-fits all framework for making technical decisions, especially when it comes to adding a new feature to a product as they often (rightly) tend to be subjective bets than scientific conclusions.
This is one opinionated decision making system (a highly simplified version of it) that tries to bring in elements of objectivity to making feature decisions in software products. This system has worked pretty well for me personally and at work, Zerodha. So, it might work for others too. View this page's source for the logic and rationale. Context: 1 2 3 4