Well, think of it this way - the standard PDF process is Very flexible. That said, they'd have to build an additional "node" for the data to be pulled. For each "node" they need to add - more complexity and risk is encountered.