Skip to main content

​
Signal Types Overview

BuildBetter can extract various signals:
  • Feature requests
  • Bug reports
  • Customer sentiment
  • Product feedback
  • Support needs
  • Integration requests

​
Basic Extraction

Automatic Signals

AI-powered detection

Manual Tagging

User-defined signals

​
Example Configurations

​
Signal Rules

# Feature Request Detection
signal_rule:
  type: "feature_request"
  triggers:
    - "need feature"
    - "would be great if"
    - "missing functionality"
  context:
    pre_words: 3
    post_words: 5
  priority:
    default: "medium"
    keywords:
      urgent: "high"
      nice: "low"

​
Sentiment Analysis

{
  "sentiment_config": {
    "granularity": "sentence",
    "aspects": [
      "product",
      "support",
      "pricing"
    ],
    "scale": {
      "range": [-1, 1],
      "neutral_threshold": 0.1
    }
  }
}

​
Real-World Examples

1

Customer Feedback

source: "support_call"
extract:
  - product_issues
  - feature_requests
  - satisfaction_level
context:
  customer_segment: "enterprise"
  product_version: "2.4.0"
2

Sales Call

source: "sales_meeting"
extract:
  - pain_points
  - competitor_mentions
  - pricing_feedback
priority: "high"
notify: ["sales_team", "product"]

​
Signal Processing

# Signal extraction from text
config = {
  "nlp": {
    "models": ["sentiment", "entity"],
    "language": "en",
    "confidence": 0.8
  },
  "output": {
    "format": "structured",
    "include_context": true
  }
}

# Signal extraction from audio
config = {
  "audio": {
    "speaker_detection": true,
    "emotion_analysis": true,
    "keyword_spotting": true
  },
  "timestamps": true
}

​
Advanced Usage

​
Custom Signal Types

# Define custom signal
custom_signal:
  name: "integration_request"
  patterns:
    - "integrate with"
    - "connection to"
    - "sync with"
  metadata:
    - platform
    - requirements
    - priority
  actions:
    - create_ticket
    - notify_team

​
Signal Correlation

Ensure signal correlation rules are properly validated to avoid false patterns.
{
  "correlation_rules": {
    "time_window": "7d",
    "min_occurrences": 3,
    "confidence": 0.85,
    "grouping": [
      "customer_segment",
      "product_area"
    ]
  }
}

​
Integration Examples

​
CRM Integration

# Salesforce signal sync
integration:
  platform: "salesforce"
  mapping:
    feature_request:
      object: "Product_Request__c"
      fields:
        description: "signal.content"
        priority: "signal.priority"
        source: "signal.meeting_id"

​
Project Management

# Jira ticket creation
automation:
  trigger: "new_signal"
  conditions:
    type: "bug_report"
    priority: "high"
  action:
    create_issue:
      project: "PROD"
      type: "Bug"
      labels: ["customer-reported"]

​
Analysis Templates

​
Trend Analysis

# Signal trend detection
analysis:
  timeframe: "30d"
  grouping:
    - signal_type
    - product_area
  metrics:
    - volume
    - sentiment
    - priority
  visualization:
    type: "trend_chart"
    breakdown: "weekly"

​
Impact Assessment

{
  "impact_scoring": {
    "factors": {
      "customer_tier": {
        "enterprise": 3,
        "business": 2,
        "starter": 1
      },
      "frequency": {
        "weight": 0.4,
        "scale": [1, 5]
      },
      "sentiment": {
        "weight": 0.3,
        "range": [-1, 1]
      }
    }
  }
}

​
Best Practices

Signal Quality

  • Validate patterns
  • Check context
  • Verify sources

Processing

  • Regular updates
  • Monitor accuracy
  • Refine rules
Regular review of signal patterns improves extraction accuracy.

​
Troubleshooting

# Troubleshooting steps
checks:
  - signal_rules
  - source_quality
  - processing_logs
  - integration_status

# Enhancement steps
improvements:
  - pattern_refinement
  - context_expansion
  - threshold_adjustment
  - validation_rules
Use test data to validate signal extraction rules before deployment.
These examples can be adapted to match your specific signal extraction needs.
⌘I