Numeric Filters on Dashboards

This is a beta feature that is still under active development. Its behavior may change in future releases.

Numeric filters let dashboard authors filter dashboard content by metric values. Use numeric filters when you want dashboard users to focus on data points that meet a metric-based condition, such as products with revenue greater than zero, customers with employee counts within a specific range, or regions where conversion rate exceeds a target.

Numeric filters work at the dashboard level. They help you avoid duplicating the same metric filter in multiple visualizations or changing the data model only to support value-range filtering.

How Numeric Filters Work

A numeric filter is based on a metric from the workspace. The filter condition is evaluated on aggregated metric values, not on raw facts. This means the filter is applied after the metric is calculated for each visualization, and also after cross-filtering and other dashboard filters are applied.

For example, if you add a dashboard numeric filter for Revenue > 1,000,000:

  • a visualization broken down by region shows only regions where total revenue is greater than 1,000,000
  • a visualization broken down by customer shows only customers where total revenue is greater than 1,000,000

The same filter can therefore affect different visualizations at the level of detail used by each visualization.

Add a Numeric Filter

To add a numeric filter to a dashboard:

  1. Open the dashboard in Edit mode.

  2. Add a filter to the filter bar.

  3. Select the metric that you want to use for filtering.

    Metrics already used on the dashboard are listed first. Other metrics available in the workspace are listed after them.

  4. Set the filter condition.

    Supported conditions include:

    • greater than
    • less than
    • equals
    • does not equal
    • range conditions

  5. Optionally, rename the filter so dashboard users see a clear, business-friendly label.

  6. Save the dashboard.

After the dashboard is saved, users can see and modify the numeric filter in the filter bar, unless the filter is configured as hidden or locked.

Default Condition

Dashboard authors can set the default condition for a numeric filter in Edit mode. The condition includes the selected operator and values. The default condition is used when users open the dashboard for the first time or reset filters to the dashboard defaults.

For example, you can create a Revenue numeric filter and set its default condition to greater than 0 so that the dashboard hides zero-revenue rows by default.

Configure Which Visualizations Are Affected

When you create a numeric filter, GoodData applies it to compatible visualizations on the dashboard.

By default, the filter is applied to visualizations that already use the selected metric or already contain a visualization-level metric filter for the same metric. Dashboard authors can also enable the filter for additional compatible visualizations.

To control whether a numeric filter affects a specific visualization:

  1. Open the dashboard in Edit mode.
  2. Open the visualization configuration.
  3. In the Filter by section, review the numeric filters listed for the visualization.
  4. Select or clear the filter as needed.
  5. Save the dashboard.

If a numeric filter cannot be applied to a visualization, the filter is disabled in the configuration. The tooltip explains that the filter is not applicable to the visualization.

Granularity

Numeric filters use the granularity of each visualization. In other words, the filter condition is evaluated at the same level of detail as the metric shown in the visualization.

For example:

  • in a revenue by region chart, the condition is evaluated for each region
  • in a revenue by customer chart, the condition is evaluated for each customer
  • in a KPI, the condition is evaluated for the KPI value

If a visualization already has a visualization-level metric filter, the dashboard numeric filter is applied together with that visualization-level filter.

Custom Granularity

Dashboard authors can also define a custom granularity for a numeric filter. Custom granularity lets you evaluate the metric at a fixed level of detail across all affected visualizations, instead of inheriting the level of detail from each visualization.

Use custom granularity when the filter should have one shared meaning across the dashboard. For example, you can configure a Revenue > 10,000 filter to be evaluated for each product category or for each day, regardless of whether an affected visualization is broken down by region, customer, or another attribute.

To configure custom granularity:

  1. Open the dashboard in Edit mode.
  2. Open the numeric filter configuration from the filter’s gear icon.
  3. In the granularity section, add one or more attributes or date dimensions.
  4. Save the filter configuration and the dashboard.

When custom granularity is set, all affected visualizations compute the filtered metric at that configured granularity. When no custom granularity is set, the default behavior is preserved and the filter uses each visualization’s own granularity.

Viewer Behavior

Dashboard viewers can change numeric filter values from the dashboard filter bar. When they change the filter, compatible visualizations and KPIs update consistently according to the selected condition.

Numeric filters work with saved views. A saved view can include the current numeric filter state, and applying the saved view restores that state.

Numeric filters also work with Apply All Filters. If the dashboard uses the Apply button, changes to numeric filters are applied together with other pending filter changes.

Drilling, Exports, Alerts, and Schedules

Numeric filters are part of the dashboard filter context. They are respected by features that use dashboard filters, including:

  • drill into dashboard
  • drill into visualization
  • drill down
  • drill to URL
  • dashboard exports
  • visualization exports
  • scheduled exports
  • alerts

When configuring drill behavior, dashboard authors can control which numeric filters are passed to the drill destination. Filters that cannot be applied to the destination are disabled in the configuration.

For drill to URL, numeric filter conditions can be passed to the custom URL using a metric filter placeholder.

Analytics as Code and Embedding

Numeric filters can be represented in Analytics as Code and included in dashboard filter groups. Developers can also control numeric filters in embedded dashboards through the dashboard filter context.

When defining numeric filters in Analytics as Code, you can include custom dimensionality in the filter definition.

Example:

- measureValueFilter:
    measure:
      identifier:
        id: "metric/order_amount"
        type: "metric"
    localIdentifier: "mvf_order_amount"
    conditions:
      - comparison:
          operator: "GREATER_THAN"
          value: 10000
    dimensionality:
      - "label/product.category"
      - "label/date.dayOfMonth"

Examples

Hide Zero-Value Results
A dashboard author adds a Revenue numeric filter and sets the default condition to greater than 0. The dashboard then shows only entities with revenue above zero.

Focus on High-Value Customers
A dashboard author adds a Revenue numeric filter and users set the condition to greater than 1,000,000. Customer-level visualizations show only customers whose aggregated revenue exceeds 1,000,000.

Filter by an Operational Range
A dashboard author adds an Employee Count numeric filter. Viewers set the condition to a range, such as 50 to 500, to focus on customers within that range.

Monitor KPI Thresholds
A dashboard author adds a Conversion Rate numeric filter. Viewers use it to show only regions where conversion rate is above a target value.

Current Limitations

The following limitations apply:

  • Ranking filters, such as Top N or Bottom N dashboard filters, are not supported.
  • Filtering on non-aggregated facts is not supported. Numeric filters are based on metrics.
  • You cannot define new metric computation logic directly in the filter. The metric must already exist in the workspace.
  • Dashboard-level numeric filters cannot be applied to visualizations that contain a ranking filter.
  • Numeric filters use the visualization granularity by default. To evaluate the filter at one shared level of detail across the dashboard, configure custom granularity for the numeric filter.
  • Numeric filters cannot be used as parent filters for dependent attribute filters.
Attribute Filters
Dependent Filters