Hi, Chris from OpenCorporates.
We're also getting the native classifications, i.e. those given by each regulators, and in a sense these are mappings. When researching this, the are many, many different classifications (particularly in the consumer finance and shadow banking area), and one of the side benefits this will surface those classifications, laying one of the foundations for a more formal ontology.
The only thing I'll add is that there are very useful classifications beyond how the regulators see it. That data are much harder to get, but very useful. For example, RIAs are a world unto themselves, but they can be reasonably segmented in a few different ways: subadvisors vs direct advisors, asset gatherers vs. asset managers, use open architecture vs. not.
