Order Surveillance: Creating an Intelligent “Net”

Preparing your firm to contend with catching spoofing, layering and other trading behavior that could run afoul of regulations, is just half the battle. While false positives or negatives can certainly be an issue, making sure that your firm is even catching the “true positives” requires some thought and planning.

Beyond just setting up systems to detect such questionable activity, trading firms need to establish the capacity, both in human resources and technologically, to catch true positives. This includes setting thresholds that make sense. “You might want to only see orders over one million shares in volume,” says Scott Burke, compliance specialist at RIMES Technologies. “If you’re only interested in what you consider to be large trades, your net should allow everything under one million shares to pass through.”

Even if the bar is set for such high-volume orders, surveillance systems must have the capacity to handle all of those orders, as Richard Levin, chair of fintech and regulation at the Polsinelli PC law firm, explains. “You have to be testing that you have sufficient capacity to handle the orders, the messaging, the throughput and the volumes that you’re planning on handling,” he says.

This also includes preventing the system from spewing too many alerts for a human reviewer to reasonably decipher and scrutinize further. A high volume of alerts creates the danger of “decision fatigue” from sorting through too many false positives. As a result, the reviewer could miss actual problematic orders or transactions in the noise.

“Surveillance teams trying to adhere to time-sensitivities may unintentionally sacrifice quality of review to handle the volume,” says Burke. “They’re missing the important stuff just to check the box that the review is complete.”

Large firms doing millions of trades need to test more frequently, but for small firms doing a limited amount of business, operating with very limited testing tools, it only makes sense to test those tools once or twice a year, according to Levin. “That’s just to make sure it’s functioning properly and that you documented that you tested it, memorialized and kept copies of the results of the test and any remediation or remedial measures that you implemented following the testing,” he says.

Catching true false positives without allowing false negatives to get by really comes down to proper tuning of the filters a firm uses to test its trades and orders. The trade data is, Burke says, like a flowing river, and filters are the fishing net. “There’s multiple variables that drive the ongoing need to tune the filter of that net,” he says. “There’s the regulatory environment. There’s your business strategy. There’s your business personnel. There’s the economic environment. That’s going to change the flow of the stream and it’s going to change ultimately what you’re interested in catching.”

Adding to those factors is the need to monitor order behavior across multiple investment products. Lists of insiders restricted from trading, to prevent front-running, must be integrated correctly with surveillance reports, Burke explains. “You want to minimize the amount of locations that you have to go to and look for your alerts,” he says.

The RIMES RegFocus Market Surveillance solution aggregates workflow alerts and manages lists of insiders that can be overlaid on surveillance data to flag trading behavior that violates regulations. The solution acts like an intelligent net that gets sticky and catches the true positives, allowing only the real positives to pass through. Learn more.


The content provided in these articles is intended solely for general information purposes, and is provided with the understanding that the authors and publishers are not herein engaged in rendering regulatory or other professional advice or services. Consequently, any use of this information should be done only in consultation with qualified legal counsel. The information in these articles was posted with reasonable care and attention. However, it is possible that some information in these articles is incomplete, incorrect, or inapplicable to particular circumstances or conditions. We do not accept liability for direct or indirect losses resulting from using, relying or acting upon information in these articles.

Posts by Topics