Using AI to analyse user behaviour patterns and improve content discovery across Oxford’s open access platforms.
Client: University of Oxford • Industry: Education • Duration: 5 months
The Challenge
Open Access Oxford needed to better understand user behavior and improve content discoverability across their digital resources. They had vast amounts of academic content but lacked insights into how users were finding and interacting with it.
Our Solution
We deployed AI-powered analytics tools to analyze user interactions and provide actionable insights for content strategy. This included behavior pattern analysis, content recommendation algorithms, and search optimization.
Results
- Enhanced Content Discovery: AI-powered search and recommendation features for better resource finding
- Data-Driven Insights: Comprehensive user behavior analysis and actionable recommendations
- Improved User Experience: Streamlined navigation and personalized content recommendations
- Effective Content Organization: Optimized content structure based on user interaction patterns
Key Metrics
45% — Search improvement: Better content discovery success rate
30% — User engagement: Increased time spent on platform
5 months — Implementation time: From analysis to full deployment
AI-Powered Content Discovery for Open Access Oxford
Project Overview
The University of Oxford’s Open Access platform hosts thousands of academic papers, research documents, and educational resources. They approached Versantus to help improve content discoverability and understand user behavior patterns across their digital resources.
The Challenge
Open Access Oxford faced several challenges with their digital content:
- Vast amounts of academic content with poor discoverability
- Limited understanding of how users interacted with content
- Ineffective content organization and categorization
- Underutilized valuable research materials
Our Solution
We implemented a comprehensive AI-powered analytics and content discovery system:
User Behavior Analysis
We deployed advanced analytics tools to:
- Track and analyze user journeys through the content
- Identify common search patterns and interests
- Understand content consumption preferences
- Map relationships between content topics
Content Recommendation Engine
We developed an intelligent recommendation system that:
- Suggested related academic papers based on viewing history
- Highlighted complementary research in different fields
- Promoted underutilized but valuable content
- Adapted to changing user interests over time
Search Optimization
We enhanced the search functionality with:
- Semantic search capabilities for academic terminology
- Topic modeling across the content database
- Faceted search for multidimensional filtering
- Search term analysis to improve metadata
The Results
The implementation delivered significant benefits:
- Enhanced content discovery with more relevant search results
- Data-driven insights into user preferences and behavior
- Improved overall user experience with intuitive navigation
- More effective content organization based on usage patterns
Long-term Impact
Beyond the immediate improvements, our solution provided Open Access Oxford with:
- A framework for ongoing content optimization
- Better understanding of their user community
- Increased utilization of valuable academic resources
- A foundation for future AI-driven enhancements