The Fraud Intelligence Cycle
By Chris Camacho
Usage of the intelligence cycle in the commercial sector is often relegated—and in many cases, exclusively so—to cyber threat intelligence (CTI) teams. But because its core principles are easily adaptable to just about any use case requiring in-depth research and complex analysis, the intelligence cycle can also be beneficial for fraud teams by adding structure to key processes and ensuring an actionable and relevant finished product.
Here’s a step-by-step walkthrough of how fraud teams can apply the intelligence cycle to support and optimize fraud investigations:
Step 1: Planning & Direction
During the first stage of the intelligence cycle, teams lay the groundwork for remainder of the process by identifying intelligence goals, setting priorities, and developing an action plan. An effective fraud investigation requires thoughtful management from the outset.
Fraud teams should begin by setting an objective and identifying a set of questions that the investigation will need to answer in order to fulfill this objective. These questions are known as intelligence requirements (IRs).
To focus on a general example, suppose the anti-fraud team at a major bank observes an uptick in account takeover (ATO) activity and consequential losses. In response, the team decides to conduct an investigation to determine how it could reduce future financial losses from ATO fraud. The team then chooses to align this objective with the following IRs:
• IR No. 1: Which tactics, techniques, and procedures (TTPs) are being used by fraudsters to access customer accounts, and what types of actors are behind this activity?
• IR No. 2: Which user behaviors are telltale signs of an account that has been hijacked by an ATO fraudster?
In addition to understanding how ATO fraudsters are gaining access to customer accounts in the first place (IR No. 1), the fraud team wants to understand how fraudsters behave once they have taken over an account so they can determine how to more effectively flag suspicious activity and freeze the account until the issue has been resolved (IR No. 2).
Step 2: Collection & Processing
In the next stage of the intelligence cycle, teams collect the information needed to comprehensively answer their IRs from a variety of sources and process it to make it useful for analysis. The information gathered during this stage will serve as the building blocks of the analysis, so it is important to leave no stone unturned when it comes to collections sources, which can include internal datasets, open-source reporting, and data from the Deep & Dark Web (DDW).
Since fraudsters tend to coordinate and/or discuss their operations on DDW forums and marketplaces, data from these illicit online communities provides a useful starting point for gaining insight into fraudster TTPs. Moreover, an organization’s internal database of user activity is an indispensable resource for analyzing threat-actor behavior during known cases of ATO.
For example, to satisfy the IRs listed in Step 1, the team could gather information from the following sources:
• DDW forum and marketplace data, information-sharing communities, open-web sources, finished intelligence reporting on fraudster TTPs.
• IR No. 2: Internal databases and user activity logs, in addition to the sources listed for IR No. 1.
Processing requirements for the above data sources could include translating discussions in foreign-language forums and reducing the volume of raw data gathered from internal databases and activity logs.
Step 3: Analysis
After collecting and processing relevant information, teams are ready for the main event: combining, integrating, evaluating, and analyzing what they’ve gathered in order to satisfy their IRs.
Returning to our example, suppose our fraud team’s analysis led to the following findings with respect to their IRs to help them determine how financial losses caused by ATO fraud can be reduced:
• IR No. 1: DDW forum data revealed Russian-speaking fraudsters were sending phishing emails linking to a malicious site designed to resemble the bank’s online portal, prompting customers to provide their username and password to resolve an urgent issue. The phishing emails were sent to a targeted list of the bank’s customers purchased on a DDW marketplace.
• IR #2: Analysis of account activity during known instances of ATO provide insight into behavioral indicators of ATO, including multiple login attempts, password resets, the use of an unrecognized device, and large amounts of money being transferred to one or more accounts belonging to other individuals.
Step 4: Production
Once their analysis is complete, teams communicate it by producing finished intelligence reports. As the primary output of the intelligence cycle, finished intelligence should be actionable and easily digestible by a wide range of stakeholders, ranging from C-level business leaders to analysts and SOC personnel at the front lines of defense.
Fraud teams often need stakeholder buy-in to take action based on their analysis. As such, to deliver return on investment and have a meaningful impact on anti-fraud efforts through their research, they must be able to effectively produce clear and concise intelligence reports that satisfy IRs.
In our example, we could summarize the content of the finished intelligence report as follows:
• IR No. 1: Russian-speaking cybercriminals are targeting customer accounts in ATO schemes that rely on sending a high volume of phishing emails, DDW forum and marketplace data, information-sharing communities, open-web sources, finished intelligence reporting on fraudster TTPs.
• IR No. 2: Telltale indicators of ATO activity include multiple login attempts, password resets, the use of an unrecognized device, and the transferring large sums of money to other accounts.
Step 5: Dissemination & Feedback
Since the overarching purpose of intelligence is to inform decisions, it is crucial that finished intelligence reports are effectively distributed to and consumed by their intended audience, which can include a wide range of stakeholders, from tactical security personnel such as SOC analysts to C-level strategic leaders such as the chief information security officer. This stage also provides stakeholders with an with an opportunity to close the feedback loop with guidance for future IRs.
Returning one last time to our example, suppose the dissemination of the fraud team’s finished intelligence report leads to the following stakeholder actions and feedback:
• To combat fraudsters’ use of malicious sites to trick users into providing their account credentials, the company implements verification measures to help customers discern decoy websites designed to resemble the online banking portal (IR No. 1).
• The CISO asks the fraud team to work with the SOC team to investigate how targeted list of customers’ email addresses ended up on a Russian-language DDW forum in the first place (IR No. 1).
• Based on the telltale indicators of ATO activity described in the report (IR No. 2), the team responsible for securing the customer portal updates the parameters for automatically freezing an account due to suspicious activity.
As fraudsters evolve their tactics to subvert existing anti-fraud controls, the intelligence cycle serves as a continuous process for answering newly uncovered questions on how to combat emerging tactics.