The other factor to consider when considering scope is cognitive load as per Team Topologies. Given that you probably don’t want (because it wouldn’t be effective) a Portfolio Management Team of 30 people, there is only so much they can consider on a quarterly basis. How much money are they trying to allocate between how many Products? How many people do they need to convene for Portfolio Reviews, etc? If thousands of people need to be within the same Portfolio because of close collaboration and dependencies, I would personally take a step back and consider the overall architecture of your solution, whether that’s a website or a submarine. It would call for more intentional architecture and less emergent architecture to better define the interfaces between the parts and therefore allow the Portfolio to be divided into sub-portfolios.