> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/useautumn/autumn/llms.txt
> Use this file to discover all available pages before exploring further.

# Analytics & Reporting

> Learn how to track usage, analyze customer behavior, and generate insights with Autumn's analytics features

Autumn's analytics dashboard provides real-time insights into how customers are using your application. Track feature usage, analyze trends, and understand customer behavior patterns.

## Accessing Analytics

There are two ways to access analytics:

1. **Global Analytics**: Navigate to the **Analytics** tab in the main sidebar to see organization-wide usage
2. **Customer Analytics**: View customer-specific usage from any customer detail page

<Note>
  Analytics requires ClickHouse to be enabled in your organization. Usage events are tracked via the Autumn SDK or API.
</Note>

## Understanding the Analytics View

The analytics page is divided into two main sections:

### Usage Chart

Visual representation of events over time:

* Bar chart showing event volume by date
* Customizable time intervals (day, week, month)
* Group by properties for segmented analysis
* Filter by event name or feature

### Events Table

Raw event data with detailed information:

* Event names and timestamps
* Customer IDs and properties
* Custom event metadata
* Pagination for large datasets

## Filtering Analytics Data

<Steps>
  <Step title="Select Time Range">
    Use the interval dropdown to choose your analysis period:

    * Last 7 days
    * Last 30 days
    * Last 90 days
    * Custom date range

    <Tip>
      Start with a 30-day view to identify overall trends, then narrow down to specific periods for detailed analysis.
    </Tip>
  </Step>

  <Step title="Choose Events to Display">
    Click the event dropdown to select which events to analyze:

    * Select specific events by name
    * Choose feature-related events
    * View all events (may be limited for performance)

    <Note>
      If no events are selected, the chart will be empty. Select at least one event to see data.
    </Note>
  </Step>

  <Step title="Set Bin Size">
    Control how data is grouped over time:

    * **Day**: Show daily usage patterns
    * **Week**: Identify weekly trends
    * **Month**: Track monthly growth
    * **Hour**: Analyze intraday patterns

    Smaller bin sizes provide more granular detail but may show more noise.
  </Step>

  <Step title="Apply Grouping (Optional)">
    Group events by custom properties to segment your analysis:

    * Customer ID: See per-customer usage
    * Custom properties: Analyze by user role, plan type, region, etc.
    * Feature IDs: Compare usage across features

    <Tip>
      Grouping creates separate bars/lines for each unique value, making it easy to compare segments.
    </Tip>
  </Step>
</Steps>

## Using Group By for Segmentation

Grouping reveals insights by breaking down events by dimensions:

<Steps>
  <Step title="Select Group By Property">
    Click the **Group By** dropdown to choose a property:

    * Properties are extracted from your event data
    * Common properties include: customer\_id, plan, region, device\_type
    * Only properties that exist in your events are available
  </Step>

  <Step title="View Grouped Chart">
    The usage chart now shows separate bars for each group:

    * Different colors represent different values
    * Legend shows all groups
    * Hover over bars to see specific values

    <Note>
      If there are too many unique values (more than 10), the chart shows only the top 10 by volume, and a warning message appears.
    </Note>
  </Step>

  <Step title="Filter to Specific Group">
    Use the group filter dropdown to focus on a single value:

    * Only shows data for the selected group
    * Useful for drilling down into specific segments
    * Clear the filter to see all groups again
  </Step>
</Steps>

## Understanding Event Properties

Events can include custom properties that provide context:

* **Standard properties**: customer\_id, timestamp, event\_name
* **Custom properties**: Any additional data you send with events
* **Nested properties**: Support for complex object structures

<Tip>
  Send consistent property names across events to make grouping and filtering more powerful. For example, always use `plan_id` instead of mixing `plan_id`, `plan`, and `subscription`.
</Tip>

## Working with the Events Table

The raw events table shows individual event records:

### Table Features

* **Pagination**: Navigate large datasets with page controls
* **Sorting**: Click column headers to sort (where enabled)
* **Property columns**: All event properties displayed as columns
* **Row details**: Click rows for more information

### Adjusting Table View

<Steps>
  <Step title="Change Page Size">
    Use the page size dropdown to show more or fewer events per page:

    * 100, 500, or 1000 events
    * Larger page sizes load more slowly but reduce pagination
  </Step>

  <Step title="Navigate Pages">
    Use the pagination controls to browse through events:

    * First/Last page buttons
    * Previous/Next page arrows
    * Current page indicator shows your position
  </Step>
</Steps>

<Note>
  The events table shows total event count at the top. Use this to understand the scope of your data.
</Note>

## Customer-Specific Analytics

View analytics for individual customers:

<Steps>
  <Step title="Open Customer Page">
    Navigate to **Customers** and select a customer.
  </Step>

  <Step title="Scroll to Analytics Section">
    The customer detail page includes a dedicated analytics section showing only that customer's events.
  </Step>

  <Step title="Analyze Customer Usage">
    Use the same filtering and grouping tools, but scoped to the customer:

    * See which features they use most
    * Identify usage trends over time
    * Understand their engagement patterns

    <Tip>
      Customer analytics are helpful for:

      * Identifying power users
      * Understanding why customers churn
      * Finding expansion opportunities
      * Debugging customer-reported issues
    </Tip>
  </Step>
</Steps>

## Common Analytics Workflows

### Tracking Feature Adoption

<Steps>
  <Step title="Filter to Feature Events">
    Select events related to a specific feature (e.g., `feature_used`, `api_call`).
  </Step>

  <Step title="Group by Customer">
    Set **Group By** to `customer_id` to see which customers use the feature.
  </Step>

  <Step title="Analyze Over Time">
    Use a 30 or 90-day interval to see adoption trends.
  </Step>
</Steps>

### Comparing Plan Usage

<Steps>
  <Step title="Set Up Grouping">
    Group events by `plan` or `plan_id` property (requires sending this with events).
  </Step>

  <Step title="Select Relevant Events">
    Choose events that indicate product usage.
  </Step>

  <Step title="Identify Patterns">
    Look for differences in usage between plan types to inform pricing strategy.
  </Step>
</Steps>

### Finding Inactive Customers

<Steps>
  <Step title="Set Long Time Range">
    Use a 30 or 90-day interval.
  </Step>

  <Step title="Review Customer List">
    In the events table, note which customer IDs are absent or have low event counts.
  </Step>

  <Step title="Take Action">
    Reach out to inactive customers or review their subscription status.
  </Step>
</Steps>

## Performance Considerations

<Note>
  Analytics queries can be resource-intensive. The dashboard implements several optimizations:

  * Event data is cached for faster repeat queries
  * Large result sets are paginated
  * Charts are limited to prevent performance issues
</Note>

<Tip>
  **Query Optimization**: Start with shorter time ranges and fewer events, then expand as needed. This keeps the dashboard responsive.
</Tip>

## Understanding Data Freshness

Analytics data updates in near real-time:

* Events appear within seconds of being tracked
* Charts refresh when you change filters
* No manual refresh required

<Tip>
  If you don't see recent events, check that your application is correctly sending events to Autumn via the SDK or API.
</Tip>

## Empty States

If you see "No events found":

1. **No events tracked yet**: Start sending events from your application
2. **Filters too restrictive**: Widen your date range or select different events
3. **ClickHouse disabled**: Contact support to enable analytics

<Note>
  The dashboard shows an empty state with a link to documentation when no analytics events exist.
</Note>

## Advanced Analytics Features

### Property Extraction

The dashboard automatically detects custom properties in your events and makes them available for grouping and filtering.

### Top Events

When you open the event selector, events are sorted by frequency, showing your most common events first.

### Data Truncation Warning

If a grouping has too many unique values (>10), you'll see a toast notification. The chart shows only the top 10 by volume to maintain performance.

## Best Practices

<Tip>
  **Consistent Event Names**: Use clear, consistent naming conventions for events (e.g., `feature_accessed`, `api_call_made`).
</Tip>

<Tip>
  **Meaningful Properties**: Send useful context with events like plan type, user role, or feature ID to enable rich segmentation.
</Tip>

<Tip>
  **Regular Monitoring**: Check analytics weekly to catch trends early and identify issues before they impact many customers.
</Tip>

<Note>
  Analytics data is stored in ClickHouse, a high-performance analytics database designed for fast queries on large datasets.
</Note>

## Exporting Data

While the dashboard doesn't have a direct export button, you can:

* Use the API to fetch raw event data
* Query ClickHouse directly if you have access
* Take screenshots of charts for reports
* Copy event data from the table view

## Next Steps

* Learn how to send events from your application using the [Autumn SDK](/sdks/javascript/installation)
* Set up custom properties for richer analytics
* Create automated reports using the API
* Monitor feature usage to inform product development

<Tip>
  Combine analytics insights with invoice data to understand the relationship between usage and revenue.
</Tip>