> ## 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.

# Filtering and Analyzing Signals

> Master advanced filtering and analysis tools to uncover insights from your signals

BuildBetter provides powerful filtering and analysis capabilities to help you find exactly what you're looking for in your sea of customer feedback. From simple keyword searches to advanced AI-powered clustering, you have the tools to uncover meaningful patterns.

## Quick Filtering

Access common filters from the main signals page:

<CardGroup cols={2}>
  <Card title="Interaction Type" icon="users">
    **Internal**: Team discussions and meetings

    **External**: Customer conversations and feedback
  </Card>

  <Card title="Extraction Method" icon="robot">
    **Automatic**: AI-extracted insights

    **Manual**: User-created signals
  </Card>

  <Card title="Time Period" icon="calendar">
    Presets: Today, 7 days, 30 days, 90 days

    Custom: Select any date range
  </Card>

  <Card title="Signal Types" icon="layer-group">
    Multi-select from available types like feature requests, bugs, complaints, praise
  </Card>
</CardGroup>

## Advanced Filtering

### Filter Panel Options

The left sidebar provides comprehensive filtering capabilities:

<Tabs>
  <Tab title="Content Filters">
    **Topics**

    * Search and select from extracted topics
    * Auto-complete suggestions as you type
    * See count of signals for each topic

    **Keywords**

    * Filter by specific terms mentioned
    * Combine multiple keywords
    * Include or exclude options

    **Tags**

    * Custom labels applied by your team
    * Hierarchical tag organization
    * Quick toggle for common tags
  </Tab>

  <Tab title="Property Filters">
    **Sentiment Range**

    * Slider from -10 (negative) to +10 (positive)
    * Filter for only negative, neutral, or positive
    * Set specific ranges for nuanced analysis

    **Severity Range**

    * Slider from -10 (trivial) to +10 (urgent)
    * Focus on critical issues or minor suggestions
    * Combine with sentiment for priority matrix

    **Bias Range** (External only)

    * Detect potentially biased feedback
    * Range from unbiased to extremely biased
    * Useful for objective analysis
  </Tab>

  <Tab title="Entity Filters">
    **Companies**

    * Filter by customer organization
    * See signal counts per company
    * Combine with revenue data for impact

    **People**

    * Individual speaker filtering
    * Track feedback from key contacts
    * Identify vocal customers

    **Personas**

    * Role-based filtering (Admin, User, etc.)
    * Understand feedback by user type
    * Custom persona definitions
  </Tab>

  <Tab title="Metadata Filters">
    **Emotions**

    * Happiness, frustration, confusion, etc.
    * Multi-select emotional states
    * Understand emotional patterns

    **Impacts**

    * Revenue, adoption, satisfaction
    * Filter by business impact type
    * Prioritize by potential effect

    **Custom Fields**

    * CRM properties
    * Account attributes
    * Any synced metadata
  </Tab>
</Tabs>

### Smart Filters (AI-Powered)

Use natural language to describe what you're looking for:

<Steps>
  <Step title="Open Smart Filter">
    Click the "Smart Filter" button or press `/` to activate
  </Step>

  <Step title="Describe Your Query">
    Type naturally: "Show me negative feedback about onboarding from enterprise customers in the last month"
  </Step>

  <Step title="Review Generated Filters">
    AI converts your description into structured filters you can further adjust
  </Step>

  <Step title="Save for Reuse">
    Save complex filter combinations as named filters for quick access
  </Step>
</Steps>

<Tip>
  Smart Filters understand context like "recent" (last 7 days), "critical" (high severity), and "upset customers" (negative sentiment).
</Tip>

## Visual Analysis Tools

### Time-Series Analysis

Track signal trends over time:

<AccordionGroup>
  <Accordion title="Line Graphs" icon="chart-line">
    * View signal volume by day, week, or month
    * Compare multiple signal types
    * Identify trending issues or improvements
    * Overlay with product releases or events
  </Accordion>

  <Accordion title="Breakdown Charts" icon="chart-bar">
    Break down signals by:

    * Type distribution
    * Sentiment analysis
    * Severity levels
    * Company or persona
    * Custom properties
  </Accordion>

  <Accordion title="Heat Maps" icon="fire">
    * Sentiment vs. Severity matrix
    * Topic frequency by time period
    * Company engagement patterns
    * Feature request clustering
  </Accordion>
</AccordionGroup>

### AI-Powered Clustering

BuildBetter's clustering feature automatically groups related signals:

**How Clustering Works**

1. AI analyzes signal content and context
2. Groups similar feedback into themes
3. Identifies unique patterns within groups
4. Provides cluster summaries and insights

**Cluster Visualization**

* **Pie Charts**: See distribution of themes
* **Drill-Down**: Click clusters to explore sub-themes
* **Breadcrumbs**: Navigate cluster hierarchy
* **Size Indicators**: Understand theme prevalence

<Note>
  Clustering works best with 50+ signals. The AI needs sufficient data to identify meaningful patterns.
</Note>

### Sentiment-Severity Matrix

The quadrant view helps prioritize actions:

```
High Severity + Negative Sentiment = Critical Issues (Top Priority)
High Severity + Positive Sentiment = Major Wins to Amplify
Low Severity + Negative Sentiment = Minor Annoyances
Low Severity + Positive Sentiment = Nice-to-Have Validations
```

## Creating Saved Views

Save complex filter combinations for quick access:

<Steps>
  <Step title="Apply Filters">
    Set up your desired filter combination
  </Step>

  <Step title="Save View">
    Click "Save Current View" and provide a name
  </Step>

  <Step title="Set Permissions">
    Choose personal or team-wide access
  </Step>

  <Step title="Quick Access">
    Access saved views from the dropdown menu
  </Step>
</Steps>

### Recommended Saved Views

<CardGroup cols={2}>
  <Card title="Weekly Review" icon="calendar-week">
    Last 7 days, all types, sorted by severity
  </Card>

  <Card title="Critical Issues" icon="exclamation-triangle">
    High severity + negative sentiment
  </Card>

  <Card title="Feature Requests" icon="lightbulb">
    Type: Feature Request, sorted by frequency
  </Card>

  <Card title="Happy Customers" icon="smile">
    Positive sentiment + praise type
  </Card>
</CardGroup>

## Analyzing Patterns

### Frequency Analysis

* **Recurring Themes**: Identify topics mentioned repeatedly
* **Trending Issues**: Spot increasing mention frequency
* **Seasonal Patterns**: Understand cyclical feedback
* **Anomaly Detection**: Flag unusual spikes or drops

### Impact Assessment

* **Revenue Impact**: Filter by high-value accounts
* **User Segment Analysis**: Compare feedback across segments
* **Feature Adoption**: Track feedback on new releases
* **Churn Indicators**: Identify at-risk patterns

### Comparative Analysis

* **Period Comparison**: This month vs. last month
* **Segment Comparison**: Enterprise vs. SMB feedback
* **Sentiment Shifts**: Track improving or declining satisfaction
* **Competition Mentions**: Monitor competitive feedback

## Exporting and Sharing

### Export Options

<Tabs>
  <Tab title="CSV Export">
    * Export filtered signals with all properties
    * Include source links and citations
    * Custom column selection
    * Bulk data for external analysis
  </Tab>

  <Tab title="Save as Saved Search">
    * Save filtered signals as a Saved Search
    * Use for document generation
    * Share with team members
    * Track changes over time
  </Tab>

  <Tab title="Generate Report">
    * Create summary document from signals
    * Include charts and visualizations
    * Add executive summary
    * Schedule recurring reports
  </Tab>
</Tabs>

## Best Practices for Analysis

<Check>
  **Start broad, then narrow**: Begin with time-based filters, then add properties to focus.
</Check>

<Check>
  **Combine filters thoughtfully**: Sentiment + Severity + Company = Actionable insights.
</Check>

<Check>
  **Save useful views**: Don't recreate complex filters—save them for the team.
</Check>

<Check>
  **Regular review cadence**: Set weekly or monthly analysis sessions using saved views.
</Check>

<Check>
  **Act on patterns, not outliers**: Focus on recurring themes rather than one-off mentions.
</Check>

## Analysis Workflows

### For Product Managers

1. Filter by feature area topics
2. Sort by severity and frequency
3. Group by company tier
4. Export for roadmap planning

### For Customer Success

1. Filter by negative sentiment
2. Add company revenue data
3. Sort by churn risk
4. Create action plans

### For Leadership

1. Use time-series for trends
2. Apply clustering for themes
3. Filter by business impact
4. Generate executive reports

<Warning>
  Remember that AI-extracted signals may need verification. Always review source material for critical decisions.
</Warning>

The filtering and analysis tools in BuildBetter transform raw feedback into actionable intelligence, helping you make data-driven decisions that truly reflect your customers' needs.