Applications to Compliance
Compliance has dramatically changed, both in scope and complexity. What used to be amenable to a simple rules-based approach is now a daunting task due not only to increased regulatory pressure, but also to ever more pervasive technology and communication media. Informal conversations about internal strategy or financial results, content posted on public platforms like Twitter, any interactions with your competitors – all of these present a risk of exposing your organization to various forms of liability.
For example, in several cases in recent years, investment banks failed to address a situation where hedge fund managers were actively promoting particular funds and promising large returns to their customers and brokers, while they privately exchanged dire outlooks on the evolution of those same funds. Because they are memorialized as electronic evidence, those contradictory statements put the company at risk of being sued by investors for misrepresenting the actual risk and the economic realities, which leads even in the best case to huge liability costs and to a damaged reputation.
This is a complex situation where context must be taken into account to distinguish between willful misrepresentation and an admissible conduct of business. Cataphora’s technology is uniquely able to detect such contradictory patterns of communication: as shown in the screenshot to the right, the most salient and anomalous emotive tones are automatically analyzed for given individuals and presented in their context, in this case specific market events and the products in question. In particular, our “two-faced” analytic identifies contradictions between messages sent to trusted colleagues and those sent to a broader audience.
Our technology uniquely and efficiently solves problems that involve massive datasets spanning large networks of individuals, and that require a deep understanding of interaction patterns and complex facts. For example:
- Compliance with antitrust regulations. By analyzing electronic data involving competitors in your industry in relation to price changes, our solutions can identify any suspicious patterns of behavior that might expose your organization to allegations of price-fixing or anti-competitive practices.
- Insider trading is still a major risk, as demonstrated by recent convictions.
- Regulatory Fair Disclosure also continues to bring additional burden on all publicly traded companies. Consider that Twitter is now recognized as a channel of dissemination of information to the market, such that tweets are subject to the SEC’s requirements for public disclosure.
Cataphora’s holistic analysis of heterogeneous data within a single, consistent model, thus provides you with reliable patterns of communication and behavior that let you identify any area of risk.
Our technology also helps you to monitor compliance with internal policies and procedures, not necessarily aligned with external regulations. For example, your organization may have large volumes of intellectual property produced by the R&D department, or other sensitive data such as customer information. That data is often heterogeneous and scattered across the network, making it challenging to detect its dissemination, whether accidentally or by malicious insiders. In addition, your enforced policy needs to comply with regulatory data retention rules, no matter how many types of business processes and communication channels are involved. Our technology thus gives you unprecedented insight into that type of information, how it is created, transmitted, and used across the organization, and provides a model of the risk carried by associated processes.
Why rules-based systems fail
Rules-based compliance systems are ill-suited in most situations, as illustrated in the following table.
|Traditional Rule-Based Systems||Using Cataphora’s Technology|
|Scope of analysis||Typically business data, mostly structured||Any type of structured or unstructured data including internal and external communication channels|
|Risk assessment||Exactly the set of pre-defined rules||You set priorities yourself and decide what matters to your organization|
|Maintenance||Costly upgrade and/or rule rewriting are necessary||The system automatically adapts to data changes|
|Relevance of results||Usually produce identifiers of breached rules which require interpretation||Alerts are always associated to real-world facts, documents, and individuals, so results are directly actionable|
|Timeliness of results||Reactive: incidents are flagged after the fact||Proactive: because indirect patterns of behavior and data management are analyzed, incidents can be anticipated|