Is the task releasable behind a feature flag (for example) ? Can it be tested independently ? Personally, I am very strict about user stories. I like them to represent a unit of customer/user value, and to be stated like that. But I think that teams can be too dogmatic about the story/task breakdown. Point taken, however. Your mileage will vary. At the end of the day, you’ll need to break down the problem into smaller problems. Whatever you call those smaller problems can be a helpful unit of progress and definition.