> ## Documentation Index
> Fetch the complete documentation index at: https://docs.buildbetter.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Signal Filtering & Properties

> Filter, search, and organize signals using advanced queries and natural language

BuildBetter provides powerful filtering capabilities to help you find and analyze the signals that matter most. Use structured filters, natural language queries, or a combination of both to explore your data.

## Filtering Overview

Filter signals by:

* Signal properties (type, sentiment, severity, emotions, business impact)
* Source information (interaction type, extraction method, data type)
* Associated entities (companies, people, personas, topics)
* CRM metadata (Salesforce, HubSpot fields)
* Time ranges and custom date periods
* Boolean logic (AND/OR/NOT) for complex queries

## Natural Language Query Builder

BuildBetter's AI-powered query builder lets you search using plain English:

### How It Works

1. Type your question in natural language
2. AI translates it to structured filters
3. Review and refine the generated query
4. Save successful queries for reuse

### Example Queries

* "Show me all bug reports from enterprise customers in the last 30 days"
* "Feature requests with high severity from this quarter"
* "Complaints about performance from companies in the healthcare industry"
* "Action items assigned to the product team this month"

<Tip>
  Start with natural language to quickly build complex filters, then switch to structured mode for fine-tuning.
</Tip>

## Structured Filtering

### Signal Property Filters

<AccordionGroup>
  <Accordion title="Type & Classification">
    **Signal Type**: Select from 35+ types (Feature Request, Bug, Complaint, etc.)

    **Interaction Type**:

    * Internal (team conversations)
    * External (customer conversations)

    **Extraction Method**:

    * Automatic (AI-detected)
    * Manual (user-created)
  </Accordion>

  <Accordion title="Sentiment & Severity">
    **Sentiment**: Filter by emotional tone (-10 to +10)

    * Very Negative: -10 to -6
    * Negative: -5 to -1
    * Neutral: 0
    * Positive: 1 to 5
    * Very Positive: 6 to 10

    **Severity**: Filter by impact level (-10 to +10)

    * Critical issues: 6 to 10
    * Important: 3 to 5
    * Moderate: 1 to 2
    * Minor: -2 to 0
  </Accordion>

  <Accordion title="Emotions & Business Impact">
    **Emotions Detected**:

    * Happiness, Frustration, Confusion
    * Excitement, Concern, Surprise
    * And more...

    **Business Impact Areas**:

    * Revenue impact
    * Adoption & usage
    * Customer satisfaction
    * Retention risk
    * Operational efficiency
  </Accordion>
</AccordionGroup>

### Entity Filters

<AccordionGroup>
  <Accordion title="Companies & People">
    **Companies**: Filter by organization name or CRM account

    **People**: Filter by contact name or participant

    **Personas**: Filter by customer segment or role
  </Accordion>

  <Accordion title="Topics & Tags">
    **Topics**: AI-detected themes and subjects

    **Keywords**: Specific terms or phrases

    **Tags**: Manual organizational labels

    **Folders**: Signals organized in collections
  </Accordion>
</AccordionGroup>

### CRM Metadata Filtering

When Salesforce or HubSpot is connected, filter by:

* **Account fields**: Industry, size, region, custom fields
* **Contact properties**: Role, status, engagement level
* **Opportunity data**: Stage, value, close date
* **Custom CRM fields**: Any field from your CRM

<Info>
  CRM metadata appears automatically in filters when integrations are active. Filter by any synced field.
</Info>

### Time Filters

**Predefined Periods**:

* Today
* Last 7 days
* Last 30 days
* Last 90 days
* This quarter
* This year

**Custom Date Ranges**:

* Select specific start and end dates
* Filter by creation date or source date
* Use relative date ranges (e.g., "last 2 weeks")

## Advanced Filtering

### Boolean Logic

Combine filters with AND/OR/NOT operators:

**Example**:

```
(Type = "Feature Request" OR Type = "Bug")
AND Sentiment < -3
AND Company Industry = "Healthcare"
```

### Nested Conditions

Create complex filter groups:

* Group related conditions
* Apply different logic per group
* Build sophisticated queries

### Saved Filter Views

* Save common filter combinations
* Name and organize your views
* Quick access to frequent analyses
* Share views with teammates

<Note>
  Saved views preserve all filter settings, column configurations, and sort orders for instant access.
</Note>

## Signal Properties Reference

### Core Properties

* **ID**: Unique signal identifier
* **Type**: Signal classification
* **Summary**: AI-generated summary
* **Content**: Full signal text/quote
* **Source**: Origin (call, conversation, import)
* **Display Date**: The date shown for the signal. You can edit this directly from the extraction table by clicking the date and using the calendar popover to pick a new date.
* **Created**: Extraction timestamp
* **Updated**: Last modification time

### Enrichment Properties

* **Sentiment**: -10 to +10 emotional score
* **Severity**: -10 to +10 impact score
* **Bias**: External signal bias (0 to 10)
* **Confidence**: Extraction confidence score
* **Emotions**: Array of detected emotions
* **Business Impact**: Impact area classifications

### Relationship Properties

* **People**: Associated contacts (with CRM data)
* **Companies**: Related organizations (with CRM data)
* **Topics**: Detected themes
* **Personas**: Customer segments
* **Tags**: User-applied labels
* **Folders**: Organization collections

### Source Context

* **Recording ID**: Link to source call
* **Timestamp**: Exact moment in recording
* **Transcript Segment**: Surrounding conversation
* **Speaker**: Who said it
* **Meeting Type**: Call classification

## Column Management

### Table View Customization

* Show/hide columns
* Reorder columns
* Adjust column widths
* Group by properties
* Sort by any column

### Available Columns

* All signal properties
* CRM metadata fields
* Custom AI-generated columns (in datasets)
* Calculated fields
* Aggregated metrics

<Tip>
  Save column configurations with your filter views for consistent table layouts.
</Tip>

## Search Capabilities

### Quick Search

* Search signal content and summaries
* Find by company or person name
* Search topics and keywords
* Filter results in real-time

### Advanced Search

* Boolean search operators
* Phrase matching ("exact match")
* Wildcard support
* Field-specific search

## Best Practices

<Check>
  **Start broad, then narrow**: Begin with type or date filters, then add specific criteria
</Check>

<Check>
  **Use natural language first**: Let AI build complex queries, then refine manually
</Check>

<Check>
  **Save frequently used filters**: Create views for common analyses
</Check>

<Check>
  **Combine entity and property filters**: Filter by "Feature Requests from Enterprise customers with high severity"
</Check>

<Check>
  **Leverage CRM metadata**: Use Salesforce/HubSpot fields to segment by business attributes
</Check>

## Troubleshooting

<AccordionGroup>
  <Accordion title="No Results Found">
    * Verify filters aren't too restrictive
    * Check date range settings
    * Ensure CRM metadata has synced
    * Try removing filters one by one
  </Accordion>

  <Accordion title="Too Many Results">
    * Add more specific filters
    * Narrow date range
    * Use AND logic instead of OR
    * Filter by higher severity/priority
  </Accordion>

  <Accordion title="CRM Fields Not Appearing">
    * Verify integration is active
    * Check sync status in Settings > Integrations
    * Ensure CRM fields are populated
    * Refresh the page to reload metadata
  </Accordion>
</AccordionGroup>

<Info>
  Filters apply in real-time. As you add or modify filters, the signal list updates immediately to show matching results.
</Info>