CDD Network Analytics Overview

[#1]

In SAS Anti-Money Laundering, CDD Network Analytics is a new optional feature as of the 2025.09 release that adds model-driven scoring and risk classification of the customer network to the customer risk ranking scoring process. It is a new, data-driven way for customer due diligence to evaluate customer risk. Instead of looking at each customer on their own, it looks at the relationships between your customers — for example, shared accounts, shared addresses, and connected transaction patterns. In this post, I will give an overview of this network analytics scoring method, including the jobs that need run to populate the CDD Networks tab in the interface.

CDD Network analytics scoring is a method in the CDD Customer Risk Ranking (CRR) scenario flow. This method adds a model, named Networks, to customer risk ranking (CRR). Networks is a quantitative scoring model that supplements the rule-driven scoring categories: Country or Region, Products, and Entity. The customers' data, demographics, and behavior information used for the model are determined on-site, and the model can be trained specifically for your organization.

The network scoring process consists of two main phases.

First, the networks model must be trained. You invoke the training manually using a batch job named cdd_network_train. By default, the decision tree model is used as the network training model in this job. Ideally, you train your model using data that is derived from feedback on CDD Networks, from investigators who evaluate CRR alerts. However, if you have never used network scoring in your CRR process, your training data might not contain this feedback. In this case, you can instead use heuristics defined by domain experts to prepare your training data.

The training job can be found in SAS Job Execution, under SAS Content, Products, SAS Anti-Money Laundering, and then jobs.

Select any image to see a larger version.
Mobile users: To view the images, select the "Full" version at the bottom of the page.

After the network model is trained, you must enable the scoring feature. To do this, in SAS Job Execution, navigate to SAS Content, Products, SAS Anti-Money Laundering, and then custom_ucmacros. Right-click the cdd_autoexec_usermode.sas file, and click Edit. Then, override the cddnet_feature_flag parameter to set the value to Y. This tells customer due diligence that you want to use network scoring in your customer risk ranking process.

The second phase is network scoring, which is seamlessly integrated into the existing cdd_prep_vi_daily job. During the network scoring process, the network model is invoked together with the other three CRR scoring methods. This job can also be found in SAS Job Execution, under SAS Content, Products, SAS Anti-Money Laundering, and then jobs.

If your organization has activated the Network Analytics scoring feature, Customer Due Diligence populates a CDD Network tab that provides information about an open customer object on the customer details page. This tab contains details about the customer's network score.

The CDD Network tab contains the following components:

• Network diagram that displays the customer's network. This includes relationships that the network generation part of the model added. This network view displays only these items for customers in the network: Accounts, Addresses, Customers, External-Party Accounts, Households, and regulatory reports that were filed.
• Network Classification Details component, which displays the most recent Network classification (High, Medium or Low risk). The classification is derived from the Network risk score (a number between 0 and 100). When you hover over the information icon, you can see the justification for the Network risk score. The Network classification date tells you the most recent date and time that the score was updated.
• Override network classification button that enables you to select a different network classification than the one calculated by the model. The next time the model re-scores the customer in the network, it will account for the override decision.
• Network Classification History shows the history of network scoring events for the customer. It includes network score override events.

To learn more information about this feature and our SAS Anti-Money Laundering solution in general, please visit https://support.sas.com/en/software/sas-anti-money-laundering-support.html

Find more articles from SAS Global Enablement and Learning here.