Advanced Filter Values

By default, filters compare a column against static values you type in. But you can also compare against other columns or connect to dashboard-level filter variables.

Value input types

Each filter condition supports three value types, selectable via buttons next to the value field:

Type	Icon	Description
Static	Data type icon (text, number, calendar, toggle)	Enter values directly
Filter Variable	Funnel icon	Connect to a filter variable
Column	Grid icon	Compare against another column

Static values

Static values are the default. You type or select the comparison value directly.

Value autocomplete

For String and Boolean columns, the input provides autocomplete based on actual values in your data. Start typing to see matches, or click the dropdown to browse.

Multiple value selection

Some operators allow selecting multiple values:

• String: Equals, Not Equal, Contains, Does Not Contain, Starts With, Ends With
• Number: Equals, Not Equal

When you select multiple values:

• “Equals” includes rows matching ANY selected value
• “Not Equal” excludes rows matching ANY selected value

Date and time values

Date columns have specialized inputs depending on the operator and time mode. See Date Filtering for details.

Null handling

Use Is Blank and Is Not Blank operators for null/empty values. These operators don’t need a value input.

On string and categorical columns, Is Blank matches empty strings as well as null and missing values.

Column comparison

Column comparison lets you compare values between two columns in the same row. Useful for finding records where column relationships matter.

How to use

1. Click the Column button (grid icon) in the value input area
2. Select the column to compare against
3. The filter compares each row’s value in the left column against the corresponding value in the right column

Data type requirements

You can only compare compatible data types:

Source column	Can compare to
String	String only
Number (int, float)	Number only
Date	Date, Datetime
Datetime	Date, Datetime
Boolean	Boolean only

Incompatible columns are disabled in the picker.

Column offset

When comparing columns, you can optionally apply an offset:

For Number columns:

• Enter a numeric offset added to the comparison column’s value
• Example: Compare revenue to cost + 100

For Date/Datetime columns:

• Enter a time offset using days, hours, and minutes
• Example: Compare due_date to created_date + 7 days

Use cases

Find overdue items:

• Column: due_date
• Operator: Before
• Compare to: current_date (if available) or use date filtering with “now”

Find profitable sales:

• Column: revenue
• Operator: Greater Than
• Compare to: cost

Find records where fields match:

• Column: shipping_address
• Operator: Equals
• Compare to: billing_address

Find records where fields differ:

• Column: assigned_to
• Operator: Not Equal
• Compare to: created_by

Dashboard date filter references

Widget filters can reference the dashboard-level date filter directly. This is useful when a widget needs a date condition that is not covered by the global date filter mapping, or when you want to use the dashboard date range as a filter value in a more complex expression.

Available references

When a filter supports date values and the dashboard has a date filter configured, the filter variable picker shows three dashboard date options under Dashboard Date Filter:

Reference	Type	Description
Dashboard Date Range	Dynamic date range	Matches dates that fall between the current start and end dates of the dashboard date filter
Dashboard Date Start	Date	Matches dates against the start date of the dashboard date filter using the chosen operator (On, Before, After, etc.)
Dashboard Date End	Date	Matches dates against the end date of the dashboard date filter using the chosen operator

How to use

1. In a widget filter, select a date or datetime column
2. Click the Filter Variable button (funnel icon)
3. Choose one of the Dashboard Date Filter options
4. Select the appropriate operator for the reference type

Fallback when the dashboard date filter is unset

If the dashboard date filter has not been set, you can choose what the condition does:

• Match all rows (default): The filter condition is ignored and all data is included
• Match no rows: The filter condition blocks all data

To change this, click the settings icon that appears after selecting a dashboard date filter reference and pick the fallback behavior.

Operator compatibility

Reference	Compatible operators
Dashboard Date Range	Equals, Not Equal
Dashboard Date Start	On, Not On, Before, After
Dashboard Date End	On, Not On, Before, After

Best practices

• Use Dashboard Date Range when you want the same date window as the global filter
• Use Dashboard Date Start or End when you only care about one boundary, such as “created after the dashboard start date”
• Set the fallback to Match no rows for strict date filtering, or Match all rows for optional filtering

Filter Variables

Filter Variables let dashboard-level filtering connect to multiple widgets. When someone changes a filter variable, all connected widget filters update automatically.

How it works

1. Create a filter variable in your dashboard (see Filter Variables)
2. In a widget’s filter, click the Filter Variable button (funnel icon)
3. Select the filter variable to connect
4. The filter’s value is now controlled by the dashboard filter variable

Benefits

• Consistent filtering: Multiple widgets filter by the same criteria
• User interactivity: Viewers can change filters without editing widgets
• Dynamic dashboards: Create flexible, reusable layouts

Compatibility

Filter variables must match the column being filtered:

• String filter variable → string column filters
• Date filter variable → date/datetime column filters
• And so on

For complete docs, see Filter Variables.

Best practices

1. Use filter variables for user-facing dashboards: Let viewers filter without edit access

2. Use column comparison for business rules: “Revenue greater than cost” is clearer than calculating margins separately

3. Match data types carefully: Mismatched types prevent filter creation or produce unexpected results

4. Consider performance: Column comparisons are evaluated row-by-row and may be slower on very large datasets

5. Document complex filters: When using multiple advanced features together, add widget descriptions explaining the filter logic