Machine Learning Survey Generation for Technical Support Pain Points

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Standardized surveys used by companies to assess user satisfaction with technical support interactions often fail to capture the specific pain points of individual users, as they may include irrelevant or inapplicable questions, leading to meaningless feedback.

Innovation Solution

A server analyzes communication sessions between users and technicians using machine learning to predict potential pain points, creating custom surveys that ask targeted questions based on the analysis, and correlates user responses to determine actual pain points, enabling manufacturers to address specific issues and improve user experience.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If standardized surveys are used to assess user satisfaction, then survey implementation is simple and consistent, but the surveys include irrelevant or inapplicable questions that do not capture specific user pain points

Engineering Contradiction:

ImproveSurvey relevance to individual user experienceVSAvoidSurvey customization complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs preliminary analysis of the communication session between user and technician before generating the survey. This includes identifying pain points, analyzing interaction quality, and determining relevant topics in advance, so that the survey is already customized and relevant when presented to the user.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The survey transitions from a static, standardized format to a dynamic, customized format. The survey questions, topics, and focus areas change based on the specific interaction analyzed, making each survey adaptive to the individual user experience rather than using a fixed template for all users.

Inventive Principle:

Principle #15Dynamics

2Measurement precision

If customized surveys are created based on individual interactions, then survey relevance and feedback quality improve, but the complexity of survey generation and processing increases

Engineering Contradiction:

ImproveFeedback accuracy regarding actual user experienceVSAvoidMachine learning model and analysis system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs self-service by automatically analyzing the communication session data, identifying pain points, and generating customized survey questions without requiring manual intervention. The machine learning model autonomously processes the interaction data and creates the appropriate survey, reducing the need for human survey designers and analysts.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual process of survey design and analysis is replaced with an automated machine learning system. Instead of humans manually creating and analyzing surveys, the system uses AI algorithms to process communication data and generate surveys, substituting mechanical human effort with automated computational processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Quantity of substance

If all potential pain points are investigated in surveys, then comprehensive feedback is obtained, but the survey becomes lengthy and includes inapplicable questions

Engineering Contradiction:

ImproveComprehensiveness of feedback coverageVSAvoidSurvey completion time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The system extracts only the relevant pain points and topics from the communication session that are actually applicable to the specific user experience. Instead of including all possible survey questions, it selectively extracts and includes only those questions that pertain to the actual interaction, removing irrelevant content.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system uses partial action by including only the necessary subset of survey questions needed to address the specific pain points identified in the interaction. Rather than exhaustively covering all potential topics, it focuses on the most relevant areas, avoiding excessive questioning while still obtaining comprehensive feedback on the actual experience.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11463328B2Training a machine learning algorithm to create survey questions

Publication Date: 2022.10.04 DELL PROD LP

AI summary

In some examples, a server may determine that a case, created to address an issue of a computing device, is closed and perform an analysis of a communication session between a user and a technician and the steps taken by the technician to resolve the issue. Machine learning may be used on results of the analysis to predict potential pain points. For example, steps that take longer than average and during which particular words spoken by the user increase in pitch and/or volume may be predicted to be potential pain points. The machine learning may create questions for inclusion in a custom survey based on the potential pain points. The custom survey may be presented to the user. The answers may be correlated with the potential pain points to determine actual pain points in the steps taken to resolve the issue.