R ... just to give you some more info. I have a complete list of field names for each API provider . These will be matched for each response, as the requests are many for each and only return a subset of names for each request. I will search for the existence of each field and map the data accordingly. A lot of the data is the response is redundant for my use. The stuff I need is mainly the delimited lists embedded in the content. I think that this approach, although more manual, will give the best performance and allow me to standardise a way of handling the many different wrappers. J.