Intrusion-detection systems (IDS), vulnerability assessment, and logging platforms have been around for a long time, being some of the very first security tools available. However, it was the inability of these technologies to adequately address specific threats that spawned new twists to these technologies.

For example, IDS was ineffective at understanding SQL queries and common application processes, so database activity monitoring (DAM) was created to fill the gap. Vulnerability assessments were fine at assessing operating system and device settings, but lack the means to understand database internal structures, so database vulnerability assessment products were built to address the need.

But it's not just the technical use cases that spawn new products;the business use cases do so as well. A great number of security products on the market are the direct result of threats against the IT infrastructure of the financial-services vertical market.

The data stolen from these institutions has direct financial value, and the finance vertical was an early adopter of new security technologies, driving the security market forward. In fact, most security solutions are geared to financial-security problems to a fault. They focus on the types of data most important to the finance vertical, work with the systems and software used in that market, and have evolved compliance solutions for regulatory requirements specific to this vertical.

That's great if your firm is in that vertical, but not so great when you manage telephony or healthcare data. I've written previously about the dynamics of innovation, but the key point is the enterprises that are the early adopters of key technologies define feature set and platform support.

Here are a couple of problems that I see:

Complex data breaks security features: The Payment Card Industry's Data Security Standard (PCI-DSS), an industry-led security standard, drives security spending for every merchant. A credit card is a very simple 16-digit number with strict formatting requirements. A patient's medical history? Not so much.

Technologies like tokenization are so focused on a simple credit card data type that the available solutions break when applied to complex data sets like patient data, medical information, insurance information, and PII. For use in the health care vertical, every vendor will need to evolve its product to handle both complex data types used in other verticals as well as adapt integration to a broader set of applications. Complex data sets may not have great value in any single element, but as aggregated data, are equally valuable. It's the same for masking technologies, which are great at obfuscating financial data sets, but struggle with complex data; they either fail to retain aggregate value, or they leak information. Lack of database support: Sybase remains a viable database platform for financial applications, and you will find a Sybase database in the IT department of just about every Wall Street firm. As such, you'll see that DAM, assessment, and auditing vendors all support that platform. Outside of the finance vertical, few use that relational database at all. Mainframe databases, Teradata, MySQL, and even NoSQL platforms are far more common for analytics in the retail, insurance, and telecom verticals, yet there are very few security solutions for these platforms.

These platforms are used in order to handle the volumes of data many analytics systems need to process, going beyond the scale of traditional relational systems. Many off-the-shelf security solutions don't support the necessary platforms, or they fall apart under normal processing loads.

Lack of application support: Solutions specific to Oracle, SAP, and other financial packages are common. The business context for the operation is understood, and advanced heuristic, content, and behavioral analysis can be applied to events.

The business use cases for the other verticals is less understood by security vendors that have not tailored their solutions to new markets. Best practices for one industry don't always map to the next. Suspicious activity around financial records is normal for patient data. Every industry has its own set of government and industry regulations it must adhere to. The point is that the common security policies in DAM, SIEM, and auditing products must be tailored to work with different use cases.

In essence, new security technologies work great for financial applications because they are the initial customers. The lack of support for alternative use cases, regulatory requirements, platforms, or data types common to other verticals leaves them wanting. That's because the deployments for most database security products are a bit of a hack -- or fail outright -- in these other verticals.

The trend I am seeing is greater adoption of security measures for PII, healthcare data, and consumer metrics. As more companies collect -- and realize value from -- these data sources, there is a growing demand for solutions that protect these other systems. Demand is still a fraction of what we see in the finance vertical, but it is growing and taking more of a lead role in defining requirements.

Adrian Lane is an analyst/CTO with Securosis LLC, an independent security consulting practice. Special to Dark Reading.