​
What are Signals?

Signals are automatically extracted insights including:
  • Customer feedback
  • Feature requests
  • Bug reports
  • Product complaints
  • Success stories
  • Integration needs

​
Signal Sources

Call Recordings

Live conversations and meetings

Documents

Uploaded files and notes

External Data

CRM, tickets, surveys

Chat Logs

Support and sales conversations

​
Extraction Process

1

Data Ingestion

Process incoming data from various sources
2

Analysis

Apply AI models for content understanding
3

Classification

Categorize and label identified signals
4

Enrichment

Add metadata and context information

​
Signal Types

​
Signal Properties

​
Automatic Properties

  • Type classification
  • Sentiment analysis
  • Priority level
  • Source reference
  • Timestamp

​
Enriched Properties

  • Customer metadata
  • Product context
  • Related signals
  • Historical data
Signal accuracy depends on data quality and proper source configuration.

​
AI Processing

NLP Analysis

Natural language understanding

Pattern Recognition

Identify trends and patterns

​
Configuration Options

​
Signal Rules

  • Detection criteria
  • Classification rules
  • Priority scoring
  • Auto-tagging rules

​
Source Settings

  • Processing frequency
  • Confidence thresholds
  • Filtering rules
  • Enrichment options
Configure signal extraction rules to match your team’s needs and priorities.

​
Integration Features

​
Data Sources

  • CRM systems
  • Help desk platforms
  • Survey tools
  • Chat platforms
  • Email systems
  • Folders (group signals for batch analysis and document generation)

​
Output Destinations

  • Project management
  • Product roadmap
  • Analytics tools
  • Team notifications
  • Folders (for organization, sharing, and document creation)
  • Dashboards (add signals to custom dashboards for visualization)

​
Best Practices

1

Source Setup

Configure and validate data sources
2

Rule Definition

Create clear extraction rules
3

Quality Check

Monitor and verify signal accuracy
4

Refinement

Adjust based on feedback and needs

​
Signal Quality

​
Quality Factors

  • Source reliability
  • Data completeness
  • Context clarity
  • Processing accuracy

​
Quality Monitoring

  • Accuracy metrics
  • False positive rates
  • Missing signals
  • Classification errors
Regularly review and adjust signal extraction rules to improve accuracy.

​
Troubleshooting

​
Automation Options

​
Scheduled Processing

  • Real-time extraction
  • Batch processing
  • Periodic updates
  • Custom schedules

​
Automated Actions

  • Signal routing
  • Notification triggers
  • Task creation
  • Report generation
Signal extraction capabilities are continuously improved through machine learning from user feedback.