A common approach to building an easy-to-use reporting system is to cut an enterprise system down to size by restricting features and providing a lot of pre-canned content. We tried that, and found it was almost, but not quite, good enough: in too many cases we had to drop down to custom code to get everything we needed, and that wiped out most of our productivity gains.

Panamplify is taking a different approach: we're building systems that know enough about the domain being reported on to operate at a higher level of abstraction. This helps us avoid the brittleness we encountered using ordinary templates. The system's knowledge is in the form of traditional artificial intelligence structures like rules, ontologies and plans. The use of these mechanisms allows the system to produce results that humans can understand and modify.