Co-browsing Field Masking and Auto-Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current co-browsing solutions lack automatic learning capabilities for webpage fields during active calls and do not allow customers to mask sensitive information from contact center agents, leading to potential privacy issues and inadequate assistance for unclear fields.
Innovation Solution
A system that enables customers to annotate and mask specific webpage fields during co-browsing sessions, storing information for generating reports and improving agent training, while preventing sensitive data from being shared with agents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If all webpage fields are displayed to the contact center agent during co-browsing, then the agent can provide comprehensive assistance, but customer privacy is compromised due to exposure of sensitive information
Solution Approach 1:
The system segments webpage fields into different categories (sensitive vs. non-sensitive) and applies different visibility rules to each segment. Customers can selectively mask specific fields while allowing others to remain visible to agents, enabling granular control over information sharing during co-browsing sessions.
Solution Approach 2:
Different visibility properties are applied to different parts of the webpage form. Each field can have its own masking status, allowing the system to protect sensitive information locally while maintaining visibility of non-sensitive fields for agent assistance.
2Reliability
If automatic learning is implemented for webpage fields during active calls, then agent training is improved, but system complexity increases
Solution Approach 1:
The system implements feedback loops where customer annotations about difficult-to-understand fields are captured during co-browsing sessions. This feedback is automatically analyzed and used to generate training reports that identify patterns and areas where agents need additional training on specific fields or concepts.
Solution Approach 2:
The system automatically collects, analyzes, and processes annotation data without requiring manual intervention. It self-generates training reports and identifies knowledge gaps, reducing the need for manual training program development and administration.
3Loss of information
If customer annotations about difficult fields are collected, then training reports are enhanced, but information processing time increases
Solution Approach 1:
The system performs preliminary processing of annotation data during the co-browsing session itself, organizing and categorizing information as it is collected. This preliminary action reduces the computational burden during report generation, allowing for faster processing of training reports when they are needed.
Data Source
AI summary
A co-browsing session between a customer communication endpoint and an agent terminal is established. User input is received, via at least one of the customer communication endpoint and the agent terminal, in the co-browsing session, which identifies a field that is difficult to understand. For example, the contact center agent may provide input that the customer does not understand a terms of use field. Information associated with the identified field that is difficult to understand is stored in a memory. The stored information is used to generate a report for the identified field. The report is then used to identify ways that may make the field more understandable by future customers. In one embodiment, the customer may also be able to mask one or more fields so that the contact center agent cannot view information in the field. For example, the customer may mask a social security number field.