Action Recommendation Engine for Software Onboarding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
New users of software applications face challenges in determining how to start using the software due to the numerous features and options available, leading to confusion and ineffective onboarding experiences.
Innovation Solution
A method and system that utilize a topic model to convert pre-signup user data and action descriptions into vectors, generating topic vectors and scores to recommend a first action to the user, thereby providing personalized onboarding guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the software application provides a large number of features and options to users, then the functionality and versatility of the software is improved, but new users experience confusion and difficulty in determining how to start using the software
Solution Approach 1:
The patent segments the overwhelming set of software features into a curated list of actionable tasks. Instead of presenting all features at once, the system divides them into discrete, manageable actions that are relevant to the user's specific context and goals, making the information digestible and actionable for new users.
Solution Approach 2:
The system performs preliminary analysis of user data (such as signup information, usage patterns, or expressed goals) before presenting actions to the user. This advance processing allows the system to pre-filter and prioritize features that are most relevant to each user, reducing the cognitive load during the onboarding process.
2Ease of operation
If the software application presents all possible actions to the user for selection, then the user has complete control over their onboarding, but users often fail to select an action or select all possible actions, making it challenging to offer meaningful guidance
Solution Approach 1:
The patent applies local quality by personalizing the action recommendations based on individual user characteristics, behaviors, and goals. Each user receives a customized subset of actions that are specifically relevant to their context, rather than a generic list. This ensures that the recommendations are neither too broad (overwhelming) nor too restrictive (limiting user choice).
Solution Approach 2:
The system incorporates feedback mechanisms to learn from user interactions with recommended actions. By analyzing which actions users take, which they ignore, and their subsequent behavior patterns, the system refines its recommendations over time, improving both the relevance of suggested actions and the overall onboarding effectiveness.
3Loss of information
If the software application provides comprehensive onboarding guidance, then user understanding is improved, but the complexity of the onboarding process increases
Solution Approach 1:
The patent applies partial action by presenting only the most relevant subset of actions to each user at any given time, rather than overwhelming them with all possible actions. This selective approach provides sufficient guidance for users to get started without presenting the entire complexity of the software's feature set upfront.
Solution Approach 2:
The system adds a new dimension to the onboarding process by incorporating contextual understanding of user goals and behaviors. Instead of organizing actions solely by software functionality, the system dimensions the presentation by user relevance, prioritizing actions that align with what the user actually wants to achieve, thereby reducing perceived complexity.
Data Source
AI summary
A method may include receiving, from a software application, pre-signup data of a user, embedding the pre-signup data to obtain a pre-signup data vector, generating, by applying a topic model to the pre-signup data vector, a pre-signup topic vector, receiving action descriptions of actions of the software application, embedding the action descriptions to obtain action description vectors, generating, by applying the topic model to the action description vectors, action description topic vectors, generating scores by multiplying the pre-signup topic vector and the action description topic vectors, and recommending, to the user and using the scores, an action.