E-discovery and auditing are flip sides of a single coin, the one concerned with retention of records and their production in litigation, the other with studying records to verify the correct of execution of corporate business processes and accounting procedures. Extending the metaphor, compliance is the coin standing on edge: neither anticipation and response to litigation (e-discovery) nor historical analysis (auditing) but rather operational rules and monitoring designed to ensure that businesses stay out of legal and accounting trouble.The three concepts — compliance, e-discovery, and auditing — are easy to confuse, but the difference is key. All three practices rely on indexing and search. All three assume retention and archival of electronically stored information (ESI). And of course e-discovery is itself the response to a mandate that calls for process compliance, for compliance with the Federal Rules of Civil Procedure, and auditing verifies compliance in e-discovery and in response to many other mandates. Yet the timings for compliance, e-discovery, and auditing efforts differ as do the scenarios in which they are applied. Nonetheless, all three could benefit from the application of knowledge discovery techniques, of text analytics and data mining, as an investigatory extension to automated processing that currently focuses almost exclusively on records retention, classification, management, search, and retrieval. Approaches to e-discovery, compliance, and auditing currently focus on content management. All three could benefit by embracing analytics.

Investigation: that would involve application of link analysis to discover relationships that go beyond participation in threaded e-mail exchanges. It would involve derivation of association rules that capture correlation between variables. It would involve information extraction — identification of named and pattern-described entities (e.g., personal, organizational, and geographic names in the first category and dates, addresses, phone numbers, and currency amounts in the second) and of facts (statements relating entities) — and record linkage to transactional data and descriptive data such as the contents of directories.

Coins have two sides and an edge, but there's also the metal they're made from, their substance. In a corporate world, compliance, e-discovery, and auditing guide us in managing information. But it's investigation, the as-yet neglected fourth dimension (so far as process automation is concerned), that reveals substance, that determines the value of particular information to corporations, regulators, and litigators.

It shouldn't be much longer before organizations understand that the technology exists to go well beyond monitoring and managing, beyond archiving and filtering, even if they're still coming to grips not only with those operational mandates but also with the basic concepts behind them. Let's understand that e-discovery is not compliance monitoring is not auditing, and then let's understand that investigatory text analytics and data mining can enhance all three, adding knowledge discovery to what is currently an operationally focused mix.E-discovery and auditing are flip sides of a single coin, the one concerned with retention of records and their production in litigation, the other with studying records to verify the correct of execution of corporate business processes and accounting procedures. Compliance is the coin standing on edge: operational rules and monitoring designed to ensure that businesses stay out of legal and accounting trouble.