Dynamic Question Wizard for Customer Support Note Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing customer support systems, including wizards and bug reporters, are inefficient in addressing unanticipated customer problems and fail to provide immediate solutions, especially when the solution already exists, and lack support for developers in creating new solutions.
Innovation Solution
A customer support wizard that dynamically selects questions based on customer data and support notes, using attribute-driven analysis to identify relevant solutions, and automatically sends data to a support note developer's system for further study if no match is found, enabling adaptive integration of new support notes and simplifying support note development.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If pre-scripted questions are used in customer support wizards, then the wizard structure is simple and easy to implement, but the system becomes inefficient when unanticipated customer problems arise and cannot identify existing solutions
Solution Approach 1:
The wizard dynamically generates questions based on customer responses rather than following a fixed pre-scripted sequence. The question generation adapts to the specific problem domain and customer answers, allowing the system to efficiently handle both anticipated and unanticipated problems while maintaining ease of implementation through automated question creation
2Adaptability or versatility
If keyword or natural language searching is used, then the system can handle unanticipated problems, but it becomes complex and may be unable to identify solutions even when they exist
Solution Approach 1:
The patent introduces an intermediary question-generation mechanism that bridges the gap between simple pre-scripted wizards and complex natural language search systems. This intermediary dynamically creates targeted questions based on customer responses and compares answers against known problem patterns, providing adaptive problem-solving without the full complexity of natural language processing
3Loss of information
If bug reporters collect trace information and submit to developers, then developers receive improved information for creating solutions, but no systematic support is provided to customers for their immediate needs
Solution Approach 1:
The system performs preliminary problem-solving by dynamically generating questions and comparing customer answers against an existing knowledge base before involving developers. This preliminary action provides immediate solutions to customers with known problems, reducing wait time, while still collecting and submitting unresolved cases to developers with enriched contextual information
4Ease of manufacture
If a fixed branching sequence is used in wizards, then the wizard is easy to implement and maintain, but it cannot adapt to different customer problems or integrate new support notes
Solution Approach 1:
The wizard employs dynamic question generation that adapts to different customer problems and new support notes without requiring manual reconfiguration of branching sequences. The system automatically generates appropriate questions based on the problem domain and customer responses, maintaining ease of maintenance through automated adaptation rather than fixed structures
Solution Approach 2:
The system changes parameters dynamically by adjusting question generation based on customer responses and problem characteristics. Rather than modifying the wizard structure, the system varies question parameters and comparison criteria to adapt to new problems and support notes, maintaining simplicity while achieving versatility
Data Source
AI summary
A customer support wizard that compares customer data with attributes of support notes to identify which notes are relevant, and to systematically shrink the number of candidate notes by asking the customer questions. Questions are selected based on an analysis of the attributes of the remaining candidate notes. If no note is found, the customer's data is sent to a development team so that a new solution can be prepared and sent back to the customer. If a customer selects a note that is not an exact match, an assumption can be made that the note will solve the customer's problem after a threshold amount of time has transpired without further customer activity. After the threshold has expired, the customer's data and the note may automatically be sent to a development team for further study.