Tableau Cloud Help

Tableau Pulse provides insights about your data based on metrics that you define. After you create a metric, you can add members of your organization as followers, and they'll receive regular email or Slack digests about their data. This digest surfaces trends, outliers, and other changes, keeping followers up to date on the data relevant to their work. To learn more about the data, these users can investigate a metric on Tableau Cloud and see how different factors contribute to changes in the data. These insights give them the information they need to make data-driven decisions without requiring them to do complex analysis in Tableau.

Metric definitions and metrics

Behind every metric in Tableau Pulse is a metric definition. Viewers interact with metrics. Metric definitions specify the core metadata for those metrics.

Metric definitions are the set of metadata that functions as the single source of truth for all the metrics based on them. Defined by a user with a Creator, Site Administrator Explorer, or Explorer (can publish) site role. The following table provides an example of the metadata captured by a metric definition.

Metric definition for Superstore Sales

Definition field	Example value
Name	Superstore Sales
Measure and aggregation	Sum of Sales
Time dimension	Order Date
Compared to	Prior year
Adjustable metric filters	Region, Category
Number format	Currency
Data source goal measure	Sales Target
Granularity of goal measure	Daily
Value going up is	Favorable

Metrics are the interactive objects that sit in front of a definition. Created when users adjust filters or time options, which means that there can be many metrics based on a definition. Users follow and explore metrics to get insights. The following tables provide an example of the options configured for metrics. These options are applied on top of the core value that is specified by the metric definition.

Metric for Superstore Sales - Technology

Metric option	Example value
Time period	Quarter to date
Filters	Category: Technology
Manual goal	80K

 

Metric for Superstore Sales - Office Supplies

Metric option	Example value
Time period	Year to date
Filters	Category: Office Supplies
Manual goal	200K

To get started in Tableau Pulse, you create a metric definition that captures the core value that you want to track. At its most basic level, this value is an aggregate measure tracked based on a time dimension. The definition also specifies options such as the dimensions that viewers are able to filter by, the way the value is formatted, and the types of insights displayed.

When you create this definition, Tableau automatically creates an initial metric and sends you to that metric's page. The initial metric created for a definition has no filters applied, but any time you or another member of your organization adjusts the metric filters or time options in a new way, Tableau Pulse creates an additional metric.

People in your organization follow metrics, not metric definitions. By following individual metrics, they get insights specific to the dimensions that matter to them. The definition exists in order to let you manage the data for metrics from a single parent object. If a field in your data source changes, you can update the definition to reflect this change, and all metrics based on that definition will also reflect the change.

Say that you're a member of a sales organization, and that organization needs to track metrics across different territories and product lines. In Tableau Pulse, you would create a metric definition that includes the core value of the sum of daily sales with adjustable metric filters for region and product line. Then, you would create metrics for each region and product line. Finally, you would add members of your organization as followers to the metrics that cover where and what they sell.

What makes Tableau Pulse different

Tableau Pulse presents a simplified way to create metric definitions, so that with only a few selections, you can make a definition that would normally require complex calculations to build in traditional Tableau viz authoring. Members of your organization use that metric definition as a jumping off point to make metrics relevant to their needs, by slicing the data based on different dimensions or time options. Because insights about these metrics are sent directly to followers, your colleagues get the data they need in their flow of work.

With Tableau Pulse, users have an easy, self-service way to take part in guided data exploration. They can ask suggested questions to see how different dimensions affect the data. This guided exploration complements the more free-form analysis that's possible with the traditional Tableau viz authoring experience and allows users unfamiliar with Tableau analysis to understand their data.

Note that though some parts of Tableau Pulse are similar to other Tableau features, Tableau Pulse combines metrics and insights in a simplified experience. In February 2024, with the release of Tableau Pulse, Ask Data and Tableau's legacy Metrics feature were retired. Like Ask Data, Tableau Pulse lets you ask questions of your data, so you can learn the how and why behind the numbers you see. Tableau Pulse also lets you create and track metrics, like the legacy Metrics feature, but Tableau Pulse metrics don’t stand alone. These metrics are the source of insights about your data.

Data source requirements for metric definitions

You create a metric definition by connecting to a published data source. For information about how to structure that data source, see Set Up Your Data for Tableau Pulse.

Configure the core definition

After making sure that your data source will work with Tableau Pulse, you're ready to start creating your definition. The definition editor is optimized for larger screens, so you should create your definition using a desktop or laptop computer rather than a mobile device.

To create metric definitions, you need a Creator, Site Administrator Explorer, or Explorer (can publish) site role on Tableau Cloud. If you have a Viewer site role, you can follow metrics and discover insights, but you can't create metric definitions.

1. From the Tableau Pulse home page, select New Metric Definition.

2. Select a data source to connect to, then select Connect.

3. For the Name, enter a name that isn't in use by other metric definitions.

   This name appears on all metrics based on the definition, so choose a name that's easy for others to understand.

4. For the Description (optional), provide brief details to help others make sense of the data.

   The description appears on the definition page. On the insights exploration page for each metric, the description shows when users select the info icon.

5. For Links (optional), add up to five links. These links appear on a metric’s insights exploration page. The links must begin with https://.

Define the metric value

1. For the Measure, select the field to track.

   You can select a measure or a dimension, but dimensions must be aggregated as a count or count (distinct) so that they result in a measure that can be tracked.

2. For the Aggregation, select how Tableau Pulse should aggregate the field you're tracking.

   If you require a more complex aggregation, see Create an advanced definition.

3. For the Temporality, select Over time or Latest point in time.

   The temporality setting controls whether the metric value and the value for the points on a metric chart are cumulative over the period or a snapshot value for the latest date. For example, a metric with a month-to-date time period and an over-time temporality would, on the 15th day of the month, show a cumulative value for days 1–15. That same metric with a point-in-time temporality would show a value that captures the data as it stands on the 15th.

   This setting also affects the options that are available for the Show sparkline values to date setting. Metrics with a point-in-time temporality always have a non-cumulative chart. In addition to the overall chart being non-cumulative, for a point-in-time temporality, each point on the metric chart shows the value for the latest date in the period covered by that point. So for a chart where each point represents a week of data, the value is for the last date in that week.

4. For Show sparkline values to date as, select whether you want the points on the metric chart to display as a running total or as non-cumulative values.

   This setting applies to the sparkline, the overview line chart, and applicable insights. For metrics that are sums, the current value shown at the top of the metric will always be a running total for the period you're tracking.

5. For the Definition filters (optional), select values to limit the metric data.

   Definition filters affect the data for all metrics based on the definition and aren't adjustable by the viewer. Fields added as definition filters change the meaning of the definition. For example, a definition filter might exclude returned orders to define net sales. If you simply want to use a field to segment the data, add an adjustable metric filter, available under the Options section. For more information, see Define metric options.

6. For the Time dimension, select the field to define the time series. In March 2025, the advanced time settings moved to the Time step. To configure those options, see Configure time settings.

How fiscal calendars work with metrics

If your time dimension is configured to use a fiscal calendar, Tableau Pulse uses that calendar. The metrics based on a definition with a fiscal calendar show fiscal years and fiscal quarters on charts and insights. When you create a definition, the fiscal start month is listed under the time dimension field, if one is set. You can't adjust the fiscal calendar in Tableau Pulse. To change it, edit the data source used by the definition. For more information, see Fiscal Dates.

Create an advanced definition

If you prefer the flexibility of working in the traditional Tableau viz authoring environment, or if you need to create calculated fields, use the advanced analytics editor. Using the advanced analytics editor is optional.

1. On the definition panel, select Create Advanced Definition.

2. Add fields to the measure, time dimension, and filters shelves.

   Only the fields, including calculated fields, that you add to these shelves are saved by the editor. Number format changes, granularity adjustments, or table calculations made in the advanced editor don't carry over. Instead, use the definition settings to set a minimum time granularity or change the number format.

3. Select Apply.

   The fields you added in the editor replace the equivalent fields in the definition panel. To edit these fields, reopen the editor. You can't edit fields configured in the advanced analytics editor in the definition panel.

Define metric options

1. For Adjustable metric filters, add at least one option.

   These filter options appear on metrics and allow users to scope the data to meet their needs. Filtering creates additional metrics from a definition. The first 20 fields that you add as adjustable metric filters also determine the dimensions used to generate insights about the data. If you add more than 20 fields, those options still appear as filters, but they aren't used to generate insights or available as breakdown options.

2. For the Number format, you can specify custom units to show for the value, or you can set the value to display as currency or as a percentage.

3. For Decimal places, select the number of decimal places to show for metric values. By default, Tableau Pulse automatically determines the number of decimal places to show. As numbers get larger, Tableau Pulse shows fewer decimal places.

Configure time settings

The time settings are optional and let you adjust the time series and comparisons for metrics so that they fit your data.

Set a date offset

A date offset changes the final point in a metric’s time series from today to a day in the past. An offset is useful if the data in the metric’s data source refreshes at a delay. Otherwise, if there isn’t any data for the final point in the time series, a metric shows a null value.

There are two types of offsets: fixed and dynamic. A fixed offset shifts the end of the time series a specified number of days. A metric with a fixed offset of 2 days shows the time period “2 Days Ago” instead of “Today.”

A dynamic date offset automatically adjusts to show the most recent date with data instead of “Today.” So if the data source doesn’t update for several days, the offset increases to continue showing the most recent date with data.

1. For the Date offset, select Fixed or Dynamic.

2. For a fixed offset, enter the number of days to shift the time series by.

Write calculated fields that work with dynamic offsets

If the measure for a metric definition uses a calculation to aggregate multiple fields, any of those fields having a null value for a date will result in the aggregate value for that date being null. A dynamic offset skips dates that have a null value to find the most recent date with data.

To make sure that Tableau Pulse doesn’t skip over dates that do have data, use an IF statement in the calculation to identify any dates that have at least one non-null field, and then apply the ZN (zero if null) function. The ZN function transforms the null values to zeroes so that they can be aggregated with the other fields that do have data, and the dynamic offset then identifies that date as having data. For more information about these functions, see Number Functions (ZN) and Logical Functions (IF).

Set the minimum granularity

The minimum granularity controls the time options available when users filter metrics. That way, users can’t select options that have no data.

1. For the Minimum time granularity, select a larger granularity if smaller units like day or week don’t make sense based on your data.

Customize the comparison

1. For Time comparison > Primary, select Prior period to compare to the prior day, week, month, quarter, or year, or select Prior year to compare to the same time range a year ago. The primary time comparison is displayed in digests and insights and on the metric overview card.

2. For Time comparison > Secondary, if you don't want metrics to show a secondary comparison, select None. The secondary comparison appears in addition to the primary comparison on a metric's insights exploration page.

Set a custom daily comparison

A custom daily comparison allows you to override the standard comparison for metrics that have a time range of “Today” or “Yesterday.” A custom comparison requires two fields in your data source: the custom daily comparison and the reference date. The reference date defines which dates show a custom comparison, and the custom daily comparison supplies the alternative dates. The fields for the reference date and the custom daily comparison must belong to the same table in the data source.

Use this setting to compare events that don’t fall on a consistent date each year. For example, you can compare Black Friday this year to Black Friday a year ago. Without a custom comparison, a metric showing data for November 28, 2025 shows a prior year comparison to November 28, 2024. With a custom comparison, you can specify a daily comparison of November 29, 2024 for the reference date of November 28, 2025. That way, Tableau Pulse compares sales data for the same event.

1. For Custom daily comparison, select a date field. This field contains the alternative dates for the daily comparison.

2. For Reference date, select a date field. This field determines which dates show a custom comparison. You can reuse the time dimension from the core definition as long as it belongs to the same data table as the custom daily comparison.

Set a custom calendar

The calendar must be defined in your data source, and all of the calendar fields must come from the same table. For more information, see Data source requirements for metric definitions.

1. Under Calendar, select Set Calendar.
2. Select the fields in your data source that define the custom calendar.
• Day of year (data type: number)
• Week of year (data type: number)
• Month of year (data type: number)
• Month name (data type: string)
• Quarter (data type: number)
• Year (data type: number)
• Reference date (data type: date)
3. Select Apply.

Configure data source goals and thresholds

Configuring a data source goal or threshold is optional. For information about structuring your data source for goals and thresholds, see Data source requirements for metric definitions.

A data source goal or threshold uses a field from your data source to show dynamic goal or threshold values on metrics. For goals, Tableau Pulse shows a metric’s progress toward the goal value. If the pace to goal insight is turned on, you also see whether the metric is expected to meet the goal by the end of the period. For thresholds, Tableau Pulse shows how far the metric is from the threshold value.

Data source goals or thresholds appear on all metrics based on a definition. The goal or threshold value is filtered by the filters applied to a metric if data exists for that level of granularity. If data doesn’t exist for specific values or dimensions, Tableau Pulse displays the goal or threshold value for a higher level of detail that does have data. That higher-level goal or threshold value might not be relevant for the filtered metric, so make sure that your data source has data of the appropriate granularity.

1. On the Goal and Threshold step, select the Type. You can configure either a goal or a threshold, but not both.

2. For Measure, select the field from your data source to use for the goal or threshold value.

3. Data source goals and thresholds inherit the definition filters from the core definition unless those filters were set in the advanced editor. If the definition filters weren’t inherited, add any relevant filters.

4. For Aggregation, select how the data is aggregated for time periods that are greater than the goal granularity.

5. For Time dimension, select the field to use for the time series.

6. For Minimum granularity, select the level of detail represented by each row of the measure. Metrics with time ranges smaller than this granularity don’t show a goal or threshold.

7. If you’re setting a threshold, for the Unfavorable Range, select whether above or below the threshold is unfavorable. The unfavorable range is shaded red on the metric chart. When the aggregate metric value crosses into this range, followers receive an alert. For more information, see Get Tableau Pulse Alerts.

8. To change the label for the goal or threshold, under Custom goal label, enter a name. This custom label appears on the goal progress indicator, chart, and in insights.

Advanced goals and thresholds

To configure a data source goal or threshold that uses calculated fields, create an advanced goal or threshold. The advanced analytics editor lets you use the familiar Tableau viz authoring environment to create calculations that define your goal or threshold.

With the advanced editor, you can write calculations using fields from your data source, or you can write calculations that define static numeric values for goals or thresholds. These static values appear on all metrics based on the definition. If you define static values, you can use logical functions to define how the values correspond to different conditions. For example, use if statements to define the goals or thresholds that appear for specific filter values. For more information about writing logical calculations, see Logical Functions.

Manual goal and threshold overrides

A manual goal or threshold is a single number that anyone with permission can set on an individual metric. If someone sets a manual goal or threshold on a metric, that value overrides any data source goal or threshold. If the manual override is removed, the data source goal or threshold appears again. To learn about manual goals and thresholds, see Set Manual Goals and Thresholds with Tableau Pulse.

Configure insights

1. Select the Insights tab.

   The fields under Insight dimensions are the same fields that you added as adjustable metric filters. Tableau Pulse uses these dimensions when monitoring your data to surface relevant insights, as shown in the insights preview.

2. For Value going up is, select whether the change is neutral, favorable, or unfavorable.

   This option controls the color for the change value: blue for neutral, green for favorable, and red for unfavorable. It also affects the language used in insights that refer to the change.

3. You can select which insights are shown for metrics by turning them on or off. Under Insight types, select Turn On or select the … menu then select Turn Off to adjust the types of insights shown. Hover over the info icon for a description of each type.
   
   Correlated Metrics: When this insight is turned on, you will need to select up to five metric definitions that can be considered for correlation with the metric you are defining. The correlated metrics insight appears only for a metric designated as a correlation candidate if the correlation coefficient between two metrics is 0.7 or greater. Additionally, the correlation must be found in the period that the metric covers or that you are asking about, in the case of Tableau Agent in Pulse.
   

   1. Select Add Candidates.

   2. In the Manage Correlation Candidates dialog box, select metric definitions that you want Tableau Pulse to watch for correlations. Try to select metric definitions of interest that don't already have an obvious causal relationship.

   3. Select Apply when you are done.

   
   Record-level Outlier: When this insight is turned on, it requires additional configuration.

   1. For the Record identifier, select a field in your data that has a unique value for every record, for example, Order ID.

   2. For the Record identifier name (optional), select a field in your data that has a name that corresponds to the record identifier, for example, Order Name.

   3. For Singular and Plural (optional), add names for the records as they should appear in insights text, for example Order and Orders.

4. Verify that the metric and insights previews look as expected, then select Save Definition.

   Tableau Pulse creates the definition along with the initial metric based on that definition, which has no adjustable metric filters applied. You can find your definition under the Browse Metrics tab on the Tableau Pulse home page.

For an overview of insight types and the insights platform, see The Insights Platform and Insight Types in Tableau Pulse.

Configure governance

The governance settings for a metric definition are optional.

Certify the definition

Metrics based on certified definitions show a badge that indicates their trusted and recommended status. Certify a metric to tell others that they can rely on that data to make business decisions.

1. Under Endorsement, turn on the Certified definition toggle.

To certify a metric definition, you must have the Create Metric Definitions permission capability for the data source that the definition is connected to. Certifying a metric definition is a separate action from certifying a data source, and data source certification doesn’t appear on metrics. To learn about certified data sources, see Use Certification to Help Users Find Trusted Data.

Restrict editing for definitions and manual goals and thresholds

By default, any user with the correct site role can edit or delete a metric definition. All users can set, edit, or delete manual goals or thresholds for the metrics based on that definition. You can restrict these actions to specific users and groups. Tableau administrators always have permission to edit because their site role gives them access.

Users with the site role of Creator, Site Administrator Explorer, or Explorer (can publish) can edit metric definitions. If you give a Viewer access to edit a definition, that user won’t be able to edit the definition because the site role doesn’t allow it. The ability to edit manual goals and thresholds isn't limited by site role.

1. Under Edit definition or Edit manual goals and thresholds, select Anyone Can Edit.

2. Select the Restrict editing toggle.

3. Enter the users and groups to add as editors.

4. Select Add.

5. Close the Manage Editors dialog.

Create metrics

After you create your definition, you’ll be taken to the initial metric for that definition. This page is the insights exploration page for that metric. On it, you can see insights based on dimensions that you select, and you can create more metrics by adjusting filters.

1. On a metric for your definition, select Filter. The filter labels become interactive.

2. Select the buttons to change the time and filter options.

3. Select the check mark button. If a metric with that combination of filters doesn’t yet exist, Tableau Pulse creates one.

To learn how viewers interact with these metrics, see Explore Metrics with Tableau Pulse.

Why is the chart for a metric incomplete?

When you create metrics with different time periods, it can reveal issues with your data. If there’s missing data for the period that the chart covers, then the metric value or the expected range might appear incomplete.

The metric value is incomplete or missing

If the metric value is incomplete, it might indicate an issue with the data, or it might be a normal consequence of the time period that the chart covers. At the beginning of the period for a period-to-date metric, the chart shows only a single point of data. If you instead want to see the most recent 7, 30, or 90 days of data, filter to show a rolling time window. If there’s no data for the current period but there’s data for previous periods, then the data source that the metric is connected to might refresh at a delay. You can set a date offset to account for this delay.

The expected range is incomplete

The expected range appears for points on the chart that have data. On non-cumulative charts, both the metric value and the expected range are missing for points without data. For metrics that show a running total, if there’s data for at least two points in the period, the chart shows a cumulative line for the metric value, but the points without data don’t have an expected range.

For metrics with a quarter-to-date time range, the quarter might start in the middle of a week. If the first week of the quarter is a partial week, then an expected range isn’t shown for that week.

Manage followers

For information about adding and removing followers from metrics, see Follow Metrics and Manage Followers in Tableau Pulse.

Edit a metric definition

If your data source changes, and the metrics that are based on it break, edit the metric definition to account for these changes. Any changes that you make to the definition will affect all metrics based on it.

1. Open a metric for the definition you want to edit.

2. Select the actions menu, then select Edit Definition.

   

How editing a definition affects metrics and goals

When you edit a definition, your changes might affect the metrics based on the definition and the goals based on those metrics.

Changed field	Affected feature	Impact
Definition filters, adjustable metric filters	Metrics	If you remove an adjustable metric filter or add a definition filter that excludes a value used in a metric filter, metrics using that filter aren’t deleted. Existing followers of an affected metric are able to adjust the filter and add other followers. However, users who aren't already followers can’t follow that metric on their own. To make it so users no longer see metrics that are based on eliminated filters, remove the followers from those metrics. Alternatively, if you want to get rid of all of the metrics for a definition, delete the definition.

Clone a metric definition

If you need to create multiple similar metric definitions, clone an existing definition. Cloning a metric definition copies all of the configuration from the original definition except for the governance step. Because you can’t have two identical metric definitions in Tableau Pulse, you must adjust the cloned definition before you save it.

1. On a metric definition page, select Clone Definition. You can also select Clone Definition from the actions (⋯) menu on a metric’s insights exploration page.
2. On the Core Definition step for the cloned definition, change at least one of the measure, aggregation, time dimension, or adjustable metric filters.
3. Review the other steps and make any adjustments needed.
4. Select Save. A cloned definition doesn’t persist until you save it.

Delete a metric definition

Deleting a metric definition also deletes all of the metrics based on it.

1. On the Tableau Pulse home page, select the Browse Metrics tab.

2. On the metric definition that you want to delete, select the actions (...) menu, then select Delete.

Add metrics to dashboards and create metrics from dashboards

For information about the Pulse metric dashboard object and recommendations for metrics based on dashboard data, see Work with Tableau Pulse Metrics and Dashboards.

Embed metrics

You can use the Embedding API to embed Tableau Pulse metrics in web pages. For more information, see Embed Tableau Pulse(Link opens in a new window).