Skip to main content

Dashboard data - Filtering & Drilling

DronaHQ's Dashboard Filters provide a way to manage data filtering. From any dashboard, you can expand the Filters sidebar to effortlessly apply multiple filters across all charts within that dashboard.

The "Value" filters allow you to specify values for any column in your dataset, offering great flexibility with various filter options. "Numerical range" filters let you easily define a range for numeric values, such as pricing or years. Additionally, Dashboard includes several time-based filters—Time Range, Time Column, and Time Grain—that enable you to filter all charts according to the time column in your dataset.

With both basic and advanced configuration settings, you can enhance your filter options and improve usability.

Add/Edit Filter
Add/Edit Filter

Different Filter Types

In the Add and edit filters window, under the Settings tab, there are several sections. The first section consists of four essential fields:

  1. Filter Type: Choose a filter type from the available options, which include Value, Numerical Range, Time Range, Time Column, and Time Grain (detailed descriptions provided below).

  2. Filter Name: Provide a brief, descriptive name for the new filter. This name usually reflects the choice made in the following Column field.

  3. Dataset: Select the dataset that will be used for the charts within the dashboard. By default, all datasets linked to the dashboard's charts are available for selection. This option appears when the Value, Numerical Range, Time Column, or Time Grain filter types are chosen.

  4. Column: Choose the column to which the filter will apply. This option is available when either the Value or Numerical Range filter types are selected.

Now, let's focus into the four filter types available in the Filter Type field.

Value Type

This filter type generates a drop-down menu that displays all available values from the selected column (as specified in the Column field).

For example, if you select the Value filter type and then choose the "Department" column, a drop-down filter menu labeled "Department" will appear on the dashboard. This menu allows you to filter the data displayed in all charts by the selected department. When the filter is applied, only the data relevant to the chosen department will be visible across all charts in the dashboard.

Filter Type - value
Filter Type - value
Filter Type - value

In this case, if you select a specific department, the dashboard will update to show data only for that department after you apply the filter.

Numerical Range

This filter type generates a range slider that allows you to define a start and end point based on the numerical values in the selected column.

For example, if you choose the "Numerical range" filter type and then select the "Electricity_Usage" column, a range slider will appear. This slider enables you to set a specific range of electricity usage. Once applied, all charts within the dashboard will display data corresponding only to the selected electricity usage range.

Filter Type - Numerical Range
Filter Type - Numerical Range
Filter Type - Numerical Range

In this case, if you adjust the slider to a particular range, the dashboard will update to show data only within that range after applying the filter.

Time Range

This filter type generates a button that, when clicked, opens the Edit Time Range functionality. This feature is the same as the Time Range option used when creating a new chart.

Filter Type - Time Range
Filter Type - Time Range
Filter Type - time Range

Time Column

This filter type generates a drop-down menu that allows you to select a new dashboard-level time column, which will override any existing time columns defined at the chart level.

tip

When should you use this filter type? The Time Column filter is ideal when your dataset contains multiple time columns. If your dataset only has one time column (which is automatically selected), this filter type may not be necessary.

Filter Type - Time Column
Filter Type - Time Column

Time Grain

This filter type creates a drop-down that enables you to select a time granularity. When applied, all timestamps are grouped based on the selected time grain.

Filter Type - Time Grain
Filter Type - Time Grain
NOTE

In the Scoping tab, there are two radio options, to determine the scope of the filter we are creating. User can select between:

  1. Apply to all panels - This will make sure that all panels with this column will be affected by this filter.
  2. Apply to specific panels - With this option only selected panels will be affected by this filter.

Parent - Child Filter

By using the Values are dependent on other filters configuration option we can apply a parent-child type relationship between filters. This configuration turns an existing filter into a sub-filter of a parent.

FeatureValueNumerical RangeTime RangeTime ColumnTime Grain
Parent-child
Pre - Filter available values
Sort filter values
Single value

Values are dependent on other filters

 Values are dependednt on other filters - Configuration
Values are dependednt on other filters - Configuration

let's have a closer look at how the configuration option works. This option is used to turn a filter into a sub-filter of an existing filter. In this example, we will make the building_Type filter a sub-filter of the department_type filter (i.e., the department type filter becomes the parent).

The idea is to allow users to first select a department, and then refine their selection further by choosing a building type within the selected department.

To start, select the building_Type filter and, in the Filter Configuration panel, check the Values Are Dependent On Other Filters option.

After doing this, a Values Dependent On drop-down menu appears — now go ahead and select department_type as the parent filter and then click Save.


Pre - Filter Data

What if you want to limit the available values in a filter based on specific criteria before users interact with it? That's where the Pre - Filter Available Values option comes in.

 Pre - Filter available values - Configuration
Pre - Filter available values - Configuration

This option is used to limit the available values in a filter based on a pre-defined condition. In this example, we will apply a pre-filter to the electricity_usage filter, restricting the available values based on the selection of department_type.

The idea is to allow users to only see electricity usage data for specific departments, such as Housing, Library, Police, and City Clerk, as defined by the pre-filter.

To start, select the electricity_usage filter and, in the Filter Configuration panel, check the Pre-Filter Available Values option.

After doing this, a Pre-Filter field appears — now go ahead and select Department as the field, set the condition to IN, and choose departments you want to include, such as Housing, Library, Police, and City Clerk. Then click Save.

NOTE

We can add multiple Pre-Filter(s), for a single filter creation. Also can write Custom SQL for filter configuration.

 Pre-filter available values - Configuration

Sort - FIlter Data

NOTE

Option is available for these Filter Types: Value, Time Range, Time Column, Time Grain

The Sort Filter Values option allows you to organize the values in a filter either in ascending or descending order. You can also choose to sort based on a specific metric from the dataset.

To use this feature, simply check the Sort Filter Values option, then select whether the sorting should be in ascending or descending order. Additionally, you have the option to sort by a metric available in the dataset.

 Sort Filter Values - Configuration
Sort Filter Values - Configuration

Single Value data filter

NOTE Option is available for this Filter Type: Numerical Range

The Single Value option allows you to restrict the filter to select just one value from the available range. You can specify the type of single value to be chosen—whether it should be the minimum, exact, or maximum value within the filter's range.

 Single Value - Configuration
Single Value - Configuration

To use this feature, check the Single Value box, and then select one of the following options:

  • Minimum: Selects the lowest available value.
  • Exact: Allows users to choose an exact value from the available options.
  • Maximum: Selects the highest available value.

This configuration is useful when you want to enforce that only one specific value is used in the filter, ensuring clarity and precision.

Filter Settings

 Different filter setting options
Different filter setting options

Here's the table with the available settings for each filter type:

SettingValueNumerical RangeTime RangeTime ColumnTime Grain
Description
Default values
Value required
Select first value by default
Select multiple values
Dynamic search
Inverse selection

Description

The "Description" setting allows you to add explanatory text or labels to filters, helping users understand the purpose or function of each filter. This can enhance usability by providing context or guidance on what kind of values are expected or what the filter will accomplish.

Filter has Default Value(s)

The Filter has default value setting enables you to pre-select a default value for a filter when it is first applied. This is useful when you want certain values to be automatically chosen by the filter, saving users from having to manually select them. For example, if you have a filter for Department Type and set Human Services as a default value, users will see Human Services selected by default when they first interact with the filter.

Filter Value is Required

When the Filter value is required option is enabled, users must specify a value for the filter before they can apply it. This setting helps prevent incomplete queries and ensures that users provide necessary input, which can improve the accuracy and performance of data retrieval. It’s particularly useful for avoiding large or unintended queries that might result from missing filter criteria.

Select First Filter Value by Default

The Select first filter value by default option automatically chooses the first item in a column’s data as the default filter value. This setting is useful when you want to ensure that a default value is always selected without manual intervention. For example, if the first item in your Department Type filter is Arts, enabling this option will pre-select Arts by default, simplifying the user experience.

NOTE

The Select first filter value by default setting helps avoid accidentally triggering expensive queries on your dashboard.

For instance, if you’re managing a dataset of flight schedules that are frequently updated, this setting will automatically select the first available flight option. This prevents you from running new queries every time the data changes, saving time and resources.

Can Select Multiple Values

The Can select multiple values setting allows users to choose more than one value from the filter. This option is beneficial when you want to offer flexibility in filtering, enabling users to select several criteria at once. For example, in a Tags filter, enabling this setting would allow users to select multiple tags, such as Technology and Innovation, simultaneously.

Dynamically Search All Filter Values

caution

Please be aware that using this option may add stress to your database.

The Dynamically search all filter values option enables real-time searching within the filter values as users type. This feature is particularly useful for large datasets with many values, making it easier for users to find and select the values they need. For example, if you have a filter with hundreds of categories, enabling this setting will allow users to quickly locate the desired category by typing part of its name.

Inverse Selection

The Inverse Selection setting allows you to exclude specific values from the filter results. Instead of including certain values, this option lets you filter out the selected ones, showing all other values. For instance, if you select Arts and Convection genres with inverse selection enabled, the filter will display all genres except these two. This can be useful for scenarios where excluding a few items is more practical than selecting many.

Scoping Filters

Scoping in dashboards allows you to control how filters are applied to specific charts, offering precise management over your data views. To access scoping options, navigate to the Filter Bar, select Add/Edit Filters, and choose the filter you want to modify. From there, go to the Scoping tab.

By default, filters like

  • Time Grain apply to all charts, while
  • Value, Numerical Range, Time Column, and Time Range filters apply to charts powered by the same dataset.

Scoping enables you to extend filter effects across multiple datasets if they share a common column, such as applying a filter across energy consumption and solar usage data that share a Building_Type field.

Additionally, you can customize scoping by applying filters to specific charts or tabs. Simply select the charts you want filtered, or deselect those you want to exclude from the filtering process. This level of control ensures that your filters are tailored to display exactly the data you need in various sections of your dashboard.


Cross-Filtering Data

Cross-filtering in dashboard enables interactive filtering across multiple charts by allowing data selections from one chart to act as filters for others.

For example, when users interact with a chart, such as selecting a particular department from the "Table Building Energy Metrics by Department," that selection can be applied as a filter across all charts that are compatible with cross-filtering.

Selecting a specific department like Aviation will update the related metrics such as "Current Solar Usage" or "Peak Electricity Demand by Building Type" pie chart. This creates a synchronized view across the dashboard, highlighting data relevant to the selected department, building, or metric.

Eligible Charts for Cross-Filtering

  • Table Chart (e.g., energy usage by building)
  • Pie Chart (e.g., peak electricity demand)
  • Bar Chart (e.g., energy consumption by department)
  • Line Chart
  • Pivot Table
  • Area Chart
  • Scatter Plot
  • Smooth Line Chart
  • Stepped Line Chart
  • World Maps
  • All charts built with ECharts

Applying Cross-Filters

Applying cross-filters is intuitive. Once you select a data point—such as a row in a table or a slice in a pie chart—the cross-filter is automatically applied across the entire dashboard. For instance, selecting the Office category in the pie chart will update the table of energy metrics to display only the energy consumption data for buildings classified under Office.

In the example image, when you select Office as a building type from the pie chart, the table displaying "Building Energy Metrics" and the current solar usage figures will immediately reflect only the Office buildings, providing a focused view of data related to this category.

 Result after appying Cross - Filteration
Result after appying Cross - Filteration

Further, let's suppose you want to provide more filtering based on the department type Aviation of the building_type from the table itself, So. simply click on the Avation department type from the table and you will see the result that both of the cross-filters are implement and a new result is showcased on each chart.

 Result after appying two Cross - Filteration
Result after appying two Cross - Filteration

You can check which cross-filters are active in the dashboard filter bar.

 Result after appying two Cross - Filteration
note

Each chart can only emit one cross-filter. If you wish to apply a filter for multiple values, use the dashboard filter bar.

Disabling Cross-Filters

In some situations, you may want to prevent dashboard consumers from using cross-filtering.

To disable cross-filtering, click on the Gear icon in the dashboard filter bar, and uncheck the Enable cross-filtering box.

 Result after appying two Cross - Filteration

Drilling into Chart Data

For users with the Viewer Workspace Role and above, accessing detailed data on a chart is made easy with the Drill to Detail feature. If you don’t see this feature, you may need to contact your Workspace Admin. Note: Embedded viewers currently don’t have access to this functionality.

The Drill to Detail feature allows you to explore the underlying data or specific metrics behind a chart in your dashboard, providing a more in-depth look at the information presented.

There are two primary ways provided to use the Drill to Detail feature:


  1. Using the Drop-Down Menu
  2. Drilling into Specific Values on a Chart

Additional Tips

  • Drill to Detail will show data that is filtered based on the dashboard’s active filters, reflecting only the relevant data.
  • You can view up to 50 rows per page in the table.
  • When applicable, the table will show formatted data by default, but you can revert to the original values by adjusting the settings via the gear icon.