Skip to content

Anywhere Deposit Fraud

Multi-Layered Check Deposit Fraud Detection:

Detecting Counterfeits, Forgeries, and Alterations From Any Deposit Channel

Solution Overview: Anywhere Deposit Fraud

Anywhere Deposit Fraud is a sophisticated OrboAnywhere Module, developed specifically to identify fraudulent checks deposited in any deposit channel, most typically at the bank of first (BoFD). Anywhere Deposit Fraud utilizes a sophisticated multi-layer approach which can be integrated into existing fraud platforms, or run independently. In both cases, the module will significantly reduce deposit fraud charge-offs and modernize the environment by scoring items and transactions so funds are not immediately available for access by the fraudster.

Anywhere Deposit Fraud Use Cases

The most common types of deposit fraud at banks involving checks include:

Over,The,Shoulder,Perspective,Of,A,Bank,Teller,Reviewing,A

Scammers create fake checks that appear legitimate, often mimicking personal, business, cashier’s, or official bank checks. These checks are deposited, and victims are tricked into sending funds back to the scammer (e.g., via wire transfer, gift cards, or cryptocurrency) before the bank detects the fraud, leaving the account holder liable for the funds. Advanced printing technologies make these checks hard to detect, even for bank employees.

Checks deposited at the BoFD which are fraudulent can be considered either first or 3rd party fraud. In the case of first party fraud, the account holder making the deposit is actually the perpetrator. In the case of third party fraud, the account holder is potentially not involved in the fraudulent transaction, and is merely a conduit to getting the money in or out of the check payment system of the FI.

Fraudulent deposited checks can be very difficult to detect, even with the many solutions available on the market. Why? The majority of these checks are drawn off other financial institutions (FIs). Because there is no immediate verification available at the item level, FIs are reliant on Reg CC funds availability rules, various account monitoring tools, or blended consortium data sources to manage risk of deposits. These systems are very vendor dependent, and many are outdated for today's complex multi-channel deposit processing.

Anywhere Deposit Fraud Imagery -- No Logo -- 24001800-01

To effectively detect check fraud attempts involving a wide range of tactics and use cases, Anywhere Deposit Fraud employs a sophisticated combination of seven (7) fraud detection layers of technology. These layers combine to deliver a best-in-breed check fraud detection solution. The system also generates hundreds of data attributes, which can be utilized in downstream business intelligence and data analytics systems.

Multi-layer Deposit Fraud Detection

OrboGraph's multi-layered technology framework is a strategic approach that integrates distinct, specialized technologies to deliver exceptional accuracy and performance. Each layer is designed with dedicated analyzers, enabling precision yet flexible scoring to complex fraud use cases. By combining these layers, the system generates a highly accurate composite score. For instance, while a fraudulent item may evade detection in one layer, subsequent layers enhance the likelihood of identification, ensuring robust and reliable outcomes.
Deposit Fraud Layers FINAL

Layer 1: Infrastructure, Profiles, & Thresholds

Developing a robust multi-layered technology approach for deposit fraud detection starts with a solid foundational infrastructure. For financial institutions, this foundation is not merely a technical requirement but a critical element that ensures seamless integration and functionality of all components involved.
Components of this layer include: multi-tenant database, workflow databases, system APIs, Deposit Profile Editor for threshold management, tuning, automated management, self learning profiles, deposit fraud rules engine, and data import(s) tools. 
Account profiles are very important to monitoring account level transactions. Profiles can start broad, or optimized into specific account types. For example, a general business profile may cover standard commercial checking accounts, while a small business profile enables finely tuned parameters for this customer type. Profiles can be tuned to the account level as well, in the event a specific account has a unique fraud behavior. By establishing a well-designed infrastructure, institutions can guarantee that information flows efficiently and that risk scores are accurately calculated, paving the way for a more reliable detection system.
Female Programmer Working in Monitoring Control Room, Surrounded by Big Screens Displaying Lines of Programming Language Code. Portrait of Woman Creating a Software. Abstract Futuristic Coding Concept

Profiles can also include 3rd party complementary data. This information is imported into the system via batch processes or by API calls, and complements the data base at the account level. Examples include account balance, average daily balance, account status, and risk scores.

Once these profiles are created, financial institutions can set precise thresholds, employing both transactional and image forensic analyzers to pinpoint potential fraud more effectively. This careful calibration of thresholds significantly mitigates the number of false positives, allowing institutions to safeguard their operations while minimizing unnecessary disruptions to legitimate bank account activities.

Layer 2: Outlier - Focused Transactional Analysis

Analyst working with Business, Concept digital transformation technology strategy, Analytics and Data Management System on computer, make a report with KPI and metrics connected to database.
Transactional Analysis plays a crucial role in detecting deposit fraud by meticulously examining an account's transactions and deposit activities. This layer involves the application of sophisticated statistical analyzer models designed to identify anomalies that could indicate fraudulent activity.

Anywhere Deposit Fraud Layer 2 transactional analyzers target dollar amount and velocity activities over daily, weekly, monthly and quarterly timeframes. Other analyzers look for unusual deposit timing, duplicates, and other unique scenarios. Analyzers are combined using harmonic mean aggregation as a way to incorporate anomalistic behaviors. The result is a score that most accurately predicts deposit fraud at the item level in real-time.

Scoring of this layer is then used by the system's self-learning logic to build account-level insights. Layers 4 and 5 are complementary to this process, as targeted image analyzers are blended with transactional characteristics of each account. 
Note that depending on how an FI decides to deploy with multiple vendors, this layer of functionality can be fulfilled with a 3rd party vendor. In many cases, machine learning-oriented systems currently provide this layer of functionality. The Anywhere Deposit Fraud analyzers can either be used to complement the existing system, or the aggregator can choose to by-pass these results.  

Layer 3: Behavioral Analysis / Account Status

Behavioral analysis serves as another input in understanding and mitigating risks associated with financial transactions by importing indicators and data attributes connected to individual accounts. This approach enables financial institutions to score deposits by utilizing information such as account risk scores and status indicators, such as new, closed, or dormant. Additionally, balance information can contribute to the account health and identify potential risks associated to each check transaction. 
For financial institutions already employing effective transaction analysis systems, the behavioral analysis layer may be viewed as optional. Existing systems may provide a robust framework for evaluating transaction authenticity, allowing institutions to maintain their operational integrity and security measures. However, the integration of the behavioral analysis of Anywhere Deposit Fraud, can offer a complementary safeguard that enhances the overall efficacy of risk detection. By supplementing existing tools with behavioral insights, financial institutions can create a more holistic view of account behavior, ultimately leading to better fraud prevention strategies and improved customer trust.
Business,Personal,Viewing,Artificial,Intelligence,For,Behavioral,Analytics,On,A

Layer 4: Image Forensics

Digital looking check new fields fraud-01
Image forensics has evolved significantly in recent years. This layer utilizes the latest deep learning image processing models developed by OrboGraph's OrbNet AI Innovation Lab. Anywhere Deposit Fraud leverages this comprehensive suite of image processing analyzers as Layer 4.
Layer 4 essentially analyzes the entire check and applies a variety of optimized image processing models. This is unique to Anywhere Deposit Fraud, as few solutions use image data today in deposit detection. Forensic analyzer examples include:
  • Check Stock Validation (CSV-AI) for  counterfeit detection
  • Automated Signature Verification (ASV-AI) for forgery detection
  • Writer Verification (WV-AI) for alteration detection
  • Payee recognition and account holder name validation
  • Full check image validation and negotiability assessment
  • IQA analysis
  • Amount field recognition, CAR/LAR discrepancy and amount verification
Other fields utilized include: payor/maker, MICR, serial, endorsement, and date fields. All data can be used for positive matching based on previously cleared checks and negative matching compared to previous fraud items.

Layer 5: Check And Account Consortium Data

Consortium data is a powerful component for deposit fraud detection. By leveraging fraud data from multiple financial institutions or internal data sources, consortiums provide real-time insights into fraud patterns, including stolen checks, altered items, and fraudulent accounts. This shared intelligence allows banks to identify risky deposits faster and more accurately than relying solely on transactional specific data.

Anywhere Deposit Fraud provides consortium enablement for financial institutions, service bureaus, bankers banks, associations, and business partners, as well as correspondent banks and other aggregators. By doing so, this strategy enables data sharing and collaboration within consortium nodes. An optional consortium approach allows for integration with third-party consortium providers like AFS but can also connect multiple consortium clusters. The combination delivers a solution which provides the widest market coverage compared to a single data resource.

Internal Consortium

Digital,Looking,Image,Of,Many,Banks,Interconnecting

OrboGraph's consortium strategy is to provide an open consortium model. The concept of an open consortium builds on multiple positive and negative data sets from various providers following the concepts of open banking. There are several consortium connectors and options available with the system today.  They include:

An internal consortium utilizes transit profiling as a unique OrboGraph component enabling banks to build out profiles of previously deposited transit checks. with a comparison to previously cleared items, this method facilitates the early identification of potentially fraudulent transactions.

Consortium Connectors

As part of the "open consortium strategy", the system can connect to multiple consortiums including Advanced Fraud Solutions TrueChecks® , EWS, Valid Systems, stolen check databases from the dark web or Telegram, internal cybersecurity data feeds, behavioral analytics, or output from existing positive and negative lists.
For example, Consortium Connector #1 retrieves account and item level data in real-time including indicators of counterfeit, non-sufficient funds (NSF), closed accounts, duplicates, and other fraud history-related activity. These risk scores are blended into the composite scoring and used in the fraud review platform for returns.
Looking ahead, initiatives such as the American Bankers Association (ABA) National Check Verification System (NCVS) aim to establish a more expansive and collaborative environment for combating check fraud, fostering a united front among banks against this pervasive threat. OrboGraph is actively involved in this, as well as several initiatives in Canada. 

Layer 6: Rules Engine, Rules Creation, Continuous Improvement

Document management concept, Businesswomen check electronic documents on digital documents on virtual screen Document Management System and process automation to efficiently document paperless operate
Rules engines serve as a pivotal tool in this battle against fraudulent activities, allowing FIs to tailor their detection around specific use cases and scenarios. Each financial entity encounters unique risks, which necessitate the implementation of custom rules. The true strength of a rules engine lies in its flexibility, enabling institutions to formulate precise parameters with a focus on minimizing false positives. By utilizing Anywhere Deposit Frauds embedded rules engine, financial institutions can leverage an expansive array of data attributes to refine their detection strategies, ensuring that genuine customer transactions are not mistakenly flagged while enhancing the identification of fraudulent ones.
OrboGraph's Continuous Improvement Process (CIP) is a program which further leverages the capabilities of rules engines by adjusting scores produced within current systems as a way to streamline the detection process. Tools to help generate rules create precise rules which can have ratios of even 1 to 3. 
Note that this layer can be an optional for deployment. When large organizations leverage enterprise rules systems, the Anywhere Deposit Fraud output may flow into the existing rules system. Otherwise, the client may leverage the Orbograph tool for a very select rule set based on limited OrboGraph data attributes.

Layer 7: Fraud Review/Queues Companions & History

The goal of an efficient fraud review platform is to limit the number of false positives and streamline the functions of a fraud analyst.  Equipped with the right tools and access to comprehensive data, these professionals are instrumental in deciphering complex fraud patterns and making informed decisions on potential threats.
To be most effective in their roles, fraud analysts require supporting information in the user interface. A robust check fraud review should include queues to assess high fraud probability items as well as historical account data and companion images. Recognizing this need, Anywhere Deposit Fraud has been developed as an integrated solution that enhances existing check fraud review platforms. Furthermore, the incorporation of Explainable AI (XAI) offers analysts insights into the rationale behind each fraud alert, enriching their investigative process. This combination of technology and expert analysis not only streamlines operations but also bolsters the overall defense against check fraud, ultimately safeguarding customer trust and financial assets.
Business,Professional,Looking,At,A,Fraud,Review,Platform,On,A

Anywhere Deposit Fraud Deployment

Deploying,Fraud,Technology,For,A,Banking,System.,Make,The,Scene

Anywhere Deposit Fraud module runs independently or as a scoring engine to an existing infrastructure, supporting low, medium, and high volume clients. All modules can run on the same hardware, or can be partitioned depending on the processing windows, volumes, and real-time versus batch requirements of the client(s).

The Anywhere Deposit Fraud module is designed to support real-time requests, providing access to any deposit channel while the transition is in process. Batch processing during day 1 or day 2 can also be implemented. In most installations, real-time and batch coexist, due to complexities in getting deposit data and images from various deposit channels immediately.

Click the button below to schedule a discussion with one of our check fraud detection experts.

Anywhere Deposit Fraud Discussion