This page was last edited on July 17, 2020, at 15:52.
Comments or questions about this documentation? Contact us for support!
![]() |
Purpose: Provides a high-level description of Context Services method of identifying customers, and contrasts it with the way that UCS (without CS) does so. |
If either method produces a unique match for the incoming customer data, there is of course no problem. The differences become relevant when there are multiple matches or when there is no match.
If UCS tries to identify a customer, and receives more than one match in return:
For example, the SCXML application may send the matched profiles to the IVR, which might prompt for the customer's name (with the grammar formed by taking the names from the matched profiles). More generally, the application will prompt for additional information and use other identification keys to further isolate the customer's identity. Once a given identity is assumed, the application will often use additional information (such as the customer's ZIP code) to validate the customer's identity. In this sense, UCS/ allows for the application to distinguish between assumed and validated customer identities.
The preceding statements about how UCS (without Context Services) identifies and creates contacts apply only to the default behavior of UCS. The "Contact Identification and Creation" chapter of the eServices 8.0 User’s Guide describes ways that you can customize this default behavior. However, what you can customize is limited to 1) the contact attributes that UCS checks and the order it checks them in, and 2) whether UCS creates a new contact in the event of no match, or if it does, a minimum set of attributes that must match. In neither case does it allow the application to expand the attributes that it checks, unlike UCS/CS.