The article critiques the common use of the term "tradeoffs" in discussions about architectural decisions in software engineering. The author argues that while "tradeoffs" implies a balance of positives and negatives in decision-making, it oversimplifies complex choices. An example from Netflix illustrates this point: the shift from regional to global personalization models in 2015. This change allowed for improved recommendation quality by leveraging global learning rather than limiting insights to specific regions. The author suggests that the focus on generating diverse training data from a global model represents a more significant challenge and opportunity than the simpler task of managing regional models. The author identifies two main ideas behind the concept of tradeoffs: that all decisions have repercussions and that no solution is perfect. While acknowledging the value of recognizing these aspects, the article highlights the drawbacks of framing decisions solely through tradeoffs. It often leads to an exaggerated focus on negative consequences, which can stall progress and obscure the more impactful benefits of certain decisions. The narrative emphasizes the importance of evaluating the long-term implications of choices rather than getting bogged down in a simplistic pros and cons list. By shifting the conversation towards whether a decision upgrades existing problems, teams can foster more productive discussions and ultimately make better architectural choices.