Decision trees have historically formed the basis for some call scripting software. “If customer is able to verify his social #, move to the next step, else Verify home address & last two transactions”. Decision trees work fine for simple scenarios such as identifying customer, or going over an upgrade offer etc. But decision trees are hard if not impossible to use in helping resolve customer issues through self service.
One big reason for this difficulty is an architectural limitation of decision trees. Each node can have only one parent in a decision tree. This limitation makes it possible to model decision trees easily in various database structures, use user interface elements like tree controls to display them and so on. However, for interactive customer self service content – this creates two fundamental limitations:
The example above refers to a customer seeking help because his TV does not work. It shows how such a simple example of customer self service could not be expressed easily on a decision tree platform, but flows naturally on a decision network platform.
Yonyx has wrapped its authoring, visualization, & analytics tools around the Decision Network Architecture. Below we cover some aspects of the Yonyx Author Guided Interactive (AGI) Map View platform that relate to DNA.
Authors tap into the power of Yonyx DNA architecture by simply being able to connect any User Response node to any other Guidance Step (except immediate loop-backs). A connection may be made in forward or backward direction, and there may be multiple User Responses connected to the same Guidance Step. The User Response at the start of a connection arrow is called the parent node. The Guidance Step it is connecting to, is called the child.