Identity verification for call centers in the financial services industry

Sam Friedewald

VP of Innovation

Identity verification is an often overlooked process that call center agents undertake on every interaction they have with customers. Agents typically ask for a couple pieces of information like birth date or social security number and then verify that the answers provided match what they have in their system.

For financial organizations such as banks and credit unions, customer identity verification is required to protect sensitive financial data and prevent fraud.

As customers, we also have an interest in protecting our finances and generally appreciate the security precautions taken as long as it doesn’t impact the customer experience too much.

Despite the importance and frequency of verifying people's identity, the process in place at most organizations is informal and poorly tracked. In many banking cores, agents can see upwards of 10 different questions they can choose from and there is little data gathered from any invalid responses. In some organizations, this information is not in any system at all and is all done from memory.

It’s no surprise that when digging into which verification questions agents choose, we find that they almost always choose the same one on every interaction, and it is the easiest one for customers to answer. Those questions also turn out to be the easiest for a fraudulent actor to learn the answer to—for example, birth date.

Building a more secure customer identity verification experience

With advances in fraud techniques such as deepfakes and simswaps, there have also been advances in prevention methods to combat those. Many of our clients are looking to implement these newer solutions and typically will utilize a combination of them.

The top four methods we are seeing include:

Digital fingerprinting: Pattern recognition in engagement metadata such as device ID’s, browsers, location, etc.
Voice analysis: After gathering an initial voice sample, a callers voice can be matched automatically in future calls
Multi-factor: Initiating a text or call with a code that the customer can read back or enter
Question and answer: The tried and true method, but made more efficient by integrating it into the agent workflow

How to securely verify customer identity over the phone with Salesforce

Follow these steps to integrate an identity verification process within Salesforce, helping to enhance security, streamline the customer experience, and reduce fraud.

Step 1: Understand the Ideal Verification Process

Collaborate with your team to map out the ideal verification process from both security and customer experience perspectives. Refine the available verification methods and develop logic to determine which methods to use in various scenarios.

For example, in a question and answer scenario, avoid asking the same question to a customer twice in a row. Fraudulent actors often gather information from initial calls to prepare for subsequent attempts, so varying the questions can disrupt their efforts.

Step 2: Create a User-Friendly Experience for Agents

Once you have a clear understanding of the methods and logic, focus on creating a seamless user experience for agents. The Identity Verification feature in Salesforce, part of the Winter '23 release, can help by providing a guided workflow. This workflow presents agents with recommended or available verification methods.

Agents can utilize built-in options to retrieve data from external systems, such as voice authenticators for risk scores, or validate responses using data within Salesforce. This guided process ensures agents can efficiently move through the verification steps or proceed to the next part of the call as needed.

verifying customer identity over the phone

Step 3: Automate Record Keeping and Utilize Machine Learning

After completing the verification process, automatically create a record of the verification in the system and link it to the relevant case and customer records. This not only meets compliance requirements but also enables the use of machine learning to enhance the system.

By maintaining historical data, you can improve customer experiences and identify fraudulent activity more effectively. For instance, if a caller repeatedly fails to answer a specific question correctly, it might trigger a more thorough verification process. Conversely, recognizing patterns, such as a real customer struggling with a particular question, can help you refine your verification methods.

Identity verification for call centers in financial services

Implementing these steps will help you build a secure and efficient identity verification process within Salesforce. This approach not only saves time and reduces hassle for customers but also significantly mitigates the risk of fraud.

model, and it improves the customers trust with you.

Interested in learning more about our customer identity verification solution for financial services?

Reach out →