If, instead, you're looking to create a reusable "pipe" that presents existing information in existing datastores in a nicer way to said franchisees, then it gets a little more hairy! Specifically, while it is entirely feasible -- even easy -- to set up Prisma with a preexisting MySQL, Postgres, or Mongo instance, you are constrained to those stores today. GraphQL, as a separate stack from Prisma, might solve the challenge elegantly even if you can't fit Prisma itself in.