Yandex DataLens dashboard

Dashboards are pages or sets of pages that contain widgets. They let you group charts in one place and add explanatory text and titles.
You can create new dashboard pages using the tab element.

Display modes

DataLens lets you display the dashboard in full-screen mode. This mode hides part of the interface elements and expands the widget area on the screen.

By default, on the mobile version of the dashboard, DataLens displays the charts one after another using the following sorting rule: left-to-right, top-to-bottom. In the tab settings, you can change the order of displaying charts and selectors in the mobile version or mailing lists.

Widgets

Widgets are dashboard elements. You can create links between selectors and chart widgets.

Yandex DataLens offers the following types of widgets:

• Chart
• Selector
• Text
• Title

Chart

Visualization as a table or chart.
You can place charts anywhere on a dashboard and change their size.
Charts can be linked to selectors.

For more information about charts, see Charts.

Selector

A filter that affects query results on its linked widgets. To add a selector to a dashboard, go to Adding selectors to dashboards.
A selector can be linked to a chart or another selector.
For more information, see Link.
A selector works within a single dashboard tab.

Text

A widget represented as text. It lets you place things like a set of links or explanatory captions on a dashboard. The Markdown markup language is supported.
For more information about Markdown, see Using Markdown in DataLens.

Title

A widget represented as a heading. It lets you differentiate charts by meaning and divide them into groups within a single dashboard page.

Link

Links determine how a selector affects one or more charts or other selectors. By default, selectors are linked with charts created within a single dataset. Selectors and charts based on different datasets can also be linked manually using aliases.

You can choose an existing dataset field as the link or create your own field where you manually set the selector values. You can specify the fields to be associated through a link in the alias configuration window.

You can use links to filter the values of selectors and charts. For example, if you select a country in the first selector on the dashboard, the second selector limits the list of cities (and vice versa).

The field used by the selector to filter on must be included in the dataset used to construct a chart. Otherwise, the link will not work. For more information, see Combining Yandex DataLens data.

The link type determines how values in selectors and charts are filtered. The type is set in the links window in the dashboard edit mode.
You can use the following types of links for a pair of widgets:

• Link, for widgets that mutually affect each other.
• Incoming link, when a widget is affected by a widget.
• Outgoing link, when a widget affects another widget.
• Ignore, for widgets that mutually ignore each other.
• No link, for widgets from different datasets or widgets that can't be linked.

Some widgets can't affect other widgets. Possible types of links between widgets are shown in the table:

Affects/Affected	Selector	Charts
Selector	Link Incoming link Outgoing link Ignore	Outgoing link Ignore
Chart	Incoming link Ignore	—

An alias

An alias is a link between fields of different datasets, which determines how widgets impact each other when you link them.

Aliases let you link widgets that are based on different datasets. For example, if you create an alias using the [Region] and [Country] fields from different datasets, a selector by the [Region] field from one dataset will filter data in a chart with the [Country] field of another dataset. For widgets based on the same dataset, aliases are created automatically. Creating aliases is required when you need to:

• Establish links between selectors that are based on different datasets, so that the values of one selector filter the list of available values of another selector.

• Establish links between a selector of the Based on dataset type from one dataset and charts from another dataset, so that the selector values filter the values in the charts.

• Establish links between a selector of the Manual input type and other selectors, so that the values of one selector filter the list of available values of another selector.

You can specify an alias for a pair of linked widgets. In this case:

• Multiple aliases can be set for a single pair of widgets.
• An alias set for one pair of widgets is applied to all linked widgets that are based on the same datasets (which use the fields specified in the alias).
• An alias works within a single dashboard tab.
• To limit the use of aliases, set Ignore as your link type for the link between your widgets on the selected field. In this case, the widgets for the selected field will have no effect on the dashboard tab.

For more information about how to create aliases, see Creating an alias.

You can delete links from an alias where more than two fields are linked. Delete a field from the alias if its associated link between datasets or widgets is not required on a dashboard tab. For more information, see Deleting an alias field.

Publishing dashboards

You can grant any internet user access to a dashboard using DataLens Public. This dashboard becomes public and any user can view it without authorization.

To share a dashboard containing Yandex Metrica or AppMetrica data, use one of the following methods:

• Add a user to an organization or cloud and grant them access to a dashboard or chart.
• Export data to a Managed Service for ClickHouse cluster and use a connection to the ClickHouse database as a source. To learn more about Managed Service for ClickHouse pricing, see Pricing.

Access management

You can configure dashboard permissions. For more information, see Managing access to DataLens.

Auto-update

You can set up automatic updates of dashboard data. The update interval is specified in seconds and the minimum value is 30 seconds. The setting is global: after saving the dashboard, auto-update will work for all users that open it. Auto-update also works in the mobile version.

The following limitations apply to auto-update:

• Data is updated only for the tab opened in the browser.

• The data is updated only for the current active tab, however:

  • If the tab isn't selected as the current tab, it's not considered active and the data is not updated.
  • If the tab is selected as the current tab, but the browser is running in the background, the tab is considered active and the data is updated.

Diagnostics

If a dashboard includes many charts that take a long time to load and process data, this may slow down the performance of the entire dashboard. You can analyze a chart's performance with an inspector.

A chart Inspector is a DataLens tool that helps you troubleshoot data loading and browser rendering issues. You can use it to determine why a chart is slow or optimize data fetching and chart rendering.

To open the chart inspector, click in the top-right corner of the chart and select Inspector.

At the top of the inspector window, there is information on the speed of basic operations:

• Config resolving: Chart configuration load time.
• Data fetching: Time it takes to load data from the source.
• Execution: Time to process data received from the source.
• Rendering: Time to render data on a chart.

Values above the optimal level are highlighted (in orange or red). To get recommendations for optimizing a metric, hover over the question mark next to its name.

For some chart types, additional metrics are displayed:

• Columns / Rows: Number of table columns and rows (for Table and Pivot table charts).
• Yandex Maps API: Time to load modules required by Yandex Maps (for Map charts).

The inspector also displays source Request IDs and request traces (Trace IDs).

The Data sources section displays information about the sources used in a chart:

• Source name.
• Link to source.
• Request body in JSON format.
• SQL query text sent to source (only if the SQL query is successful and its results are rendered in the chart).