In a large pipelined, distributed architecture, there's probably dozens of consumers of the databases that all look rather similar: they do some queries and pack up some reports for internal consumption. Marketing, ab testing, user experience, etc. Yet another consumer sharing 99% of the same code might not stand out.