Arthur B said: I tried to do some estimations. Assuming that there are about 3 million users by now, and 95% are free users, then Roll20 has a theoretical storage volume of nearly 500 TB. Nearly 60% of that volume is for free users, paid by the subscribers. Now with the storage volume increased for the subscribers, the storage volume would go up to nearly 900 TB (again, theoretically, as I'm just focusing on the maximum storage needed). Considering that most Pro and Plus users will never use that additional storage, it's pretty safe to assume that in reality, Roll20 doesn't need to stock up much in terms of additional storage space. If Roll20 would now also triple the storage for the free users, the overall storage volume would jump up to 1400 TB. And opposite to the increase for Pro and Plus users, a lot of free users would definitely use their storage, forcing Roll20 to invest in additional storage - for users that don't chip in a dime. Regardless of whether storage is being used or not, having to allocate that space on the server itself will incur the same cost. For example, if you bought a physical storage unit and filled it halfway up, you would still pay the entire price for the physical space despite the capacity not being full. Given the fact they were able to increase Pro/Plus on such a quick whim does point to the notion that there is wiggle room in the budget. While we may not know if the total remaining storage could work for all free users, what *could* be done is potentially only granting space to non-bot, non-banned, or non-duplicate free accounts. Once again, the fact that of the many free users I've spoken to have done something similar points to there already being a space issue, which is now exacerbated with the loss of Fanburst (not Roll20's fault). One alternative I did see would be using something ala Google Drive or Dropbox for third party hosting to ease the server load for images and audio files. I understand the want to host everything on the site, but major platforms like those listed above have perfectly fine existing API and no chance of going under anytime soon, some for more obvious reasons. Alternatively, a lesser package above free that would only include storage space and few if none of the higher features such as scripting, fog of war or sfx. The former would help lower server costs, the latter may assist in drawing in new Game Masters with a lower tier (or players in many ongoing games).