As for actual math? I dont know the current numbers, but I do remember in 2017 when they talked about hitting 2million users. i wonder how many of them are active? because thats important too. But lets say its 2 million. Its likely more. Lets talk about 50 meg raise for 2 million users. (where it would be rare if its anything greater than 15% active users), Thats 50 terabytes. (again, id bet that number is likely bigger, as thats over a year old) For users that mostly dont use it, according to roll20. Increasing that amount of storage would be ridiculously expensive, and it would either mean more capital investment or the costs would have to go to the advertisement, or pro users. And honestly, to a certain degree, thats what many pro users might be thinking. Now i know we all have computers with terabytes, and we're thinking, "gee thats only about 15 Grand, maybe?". But on servers? with indexing the data? That can be indeed a software nightmare. For something that honestly, probably isnt needed. Again, I would go back to a request system. Rather than just unilaterally give the free users (who again, i imagine are more inactive than active, thats the reality of websites), additional storage, creating some type of system allowing them to REQUEST additional storage. Its a little more work on the roll20 folk than not doing it at all, but its a realistic compromise, and something that is more likely financially feasable. Many pro users honestly feel that they pay for the free users. And to many extents, thats often true. In many gaming groups its rare that everyone is pro. In a larger scale, its not fully true. But there is still some proof to that. You need to support your paid customers first, and then look to support your marketting reasons (the free customers). You cant completely ignore the free customers though, because without them, the paid customers wouldnt have people to play with, and thus wouldnt pay for the service. So obviously you cant ignore the free customers either. But as others have said, this just came out. You have to look at supporting the paid customers first. And then step 2 is see what the real need of the free customers is. (which is why i suggest implementing some type of storage request system).