Skip to main content

Memory Analytics

Reeflect provides comprehensive analytics to help you understand memory usage, performance, and patterns. This is especially valuable for enterprise applications where visibility into AI memory systems is critical.

Analytics Capabilities​

Usage Metrics

Track memory creation, access, and retrieval patterns.

  • Memory creation rate
  • Retrieval frequency
  • Access patterns
  • Memory importance distribution

Performance Metrics

Monitor system performance and efficiency.

  • Retrieval latency
  • Embedding generation time
  • Storage usage
  • Token utilization

Content Analytics

Analyze the content and patterns within memories.

  • Topic distribution
  • Sentiment analysis
  • Entity recognition
  • Temporal patterns

Quality Metrics

Evaluate memory system effectiveness.

  • Contradiction rate
  • Retrieval relevance
  • Memory coherence
  • Reasoning accuracy

Using the Analytics API​

from reeflect.analytics import MemoryAnalytics

# Create analytics instance
analytics = MemoryAnalytics(memory_system)

# Get basic usage statistics
usage_stats = analytics.get_usage_stats(
    namespace="user_preferences",
    time_period_days=30
)

print(f"Total memories: {usage_stats['total_memories']}")
print(f"Active memories: {usage_stats['active_memories']}")
print(f"Creation rate: {usage_stats['creation_rate_per_day']} memories/day")
print(f"Average importance: {usage_stats['avg_importance']}")

# Get retrieval performance metrics
retrieval_metrics = analytics.get_retrieval_metrics(
    namespace="user_preferences",
    time_period_days=7
)

print(f"Average retrieval time: {retrieval_metrics['avg_retrieval_time_ms']}ms")
print(f"Average relevance score: {retrieval_metrics['avg_relevance_score']}")
print(f"Cache hit rate: {retrieval_metrics['cache_hit_rate']}")

# Get content analytics
content_analytics = analytics.analyze_content(
    namespace="user_preferences"
)

print("Top topics:")
for topic, count in content_analytics['topics'].items():
    print(f"  - {topic}: {count}")

print("Sentiment distribution:")
for sentiment, percentage in content_analytics['sentiment'].items():
    print(f"  - {sentiment}: {percentage}%")

Exporting Analytics Data​

You can export analytics data for use in external visualization tools:

# Export analytics data to various formats
analytics.export_data(
    metrics=["usage", "performance", "content"],
    format="csv",  # Options: csv, json, excel
    output_path="./memory_analytics",
    time_period_days=90
)

# Export data for specific dashboard integration
analytics.export_for_dashboard(
    dashboard_type="grafana",  # Options: grafana, kibana, tableau, custom
    metrics=["usage", "performance"],
    output_path="./dashboard_data"
)
Enterprise Feature

For enterprise users, Reeflect offers a dedicated analytics dashboard with real-time monitoring, customizable alerts, and advanced visualization capabilities. See the Enterprise Setup guide for more information.

Next Steps​

Explore the Memory Graph to learn how to visualize and explore the relationships between memories as an interactive network.