How to Set Up Tableau Pulse

Tableau Pulse delivers automated metric digests and AI-generated insights to analysts and business users inside Tableau Cloud. Setting it up requires a site administrator to enable the feature, at least one published data source on the site, and users with Explorer or Creator roles to define metrics. Once configured, followers receive scheduled email or Slack digests with trend summaries, anomaly alerts, and natural-language explanations of changes, without needing to open a workbook.

What Tableau Pulse Does

Tableau Pulse sits on top of your existing Tableau Cloud data sources and creates a metrics layer that functions separately from dashboards. Instead of asking users to open a workbook, Pulse pushes key numbers directly to where people work: email inboxes, Slack channels, and the Tableau Cloud home page.

The core unit is a metric definition. An administrator or data analyst creates a definition by pointing Pulse at a published data source, selecting a measure (revenue, headcount, open tickets), and specifying how to aggregate it. Once a definition exists, any user with access to the underlying data source can follow it and receive regular digests.

In January 2026, Tableau added the ability to disable digests and alerts at the site level, giving administrators tighter control over phased rollouts. The March 2026 release introduced Favorites, which allows users to filter their Pulse home page and configure digests to show only prioritized metrics.

Requirements

Before you begin, confirm the following:

• A Tableau Cloud site with a Creator, Explorer (can publish), or Site Administrator role for the person creating metric definitions.
• At least one published data source on the site. Tableau Pulse only works with published data sources, not local extracts or desktop files.
• A Tableau+ license if you plan to use the conversational Q&A (Discover) feature or cross-metric analysis. Standard Tableau Cloud licenses cover core Pulse functionality.
• A Salesforce org with Einstein generative AI enabled, if you want AI-written insight summaries in your digests.

The AI insights are powered by the Einstein Trust Layer, which masks data values before they reach the language model and does not use customer data to train external models.

Step 1: Enable Tableau Pulse

Only a site administrator can turn on Tableau Pulse. Log in to Tableau Cloud with an administrator account and go to Settings from the left navigation menu. Scroll to the Tableau Pulse section.

Under Deployment, click "Turn on Tableau Pulse." You will see two options: enable it for all users on the site, or restrict it to a specific group. For an initial rollout, selecting a single user group is a practical way to test the configuration with a smaller audience before expanding site-wide.

After saving, a Pulse option will appear in the main navigation for eligible users.

Step 2: Prepare a Published Data Source

Tableau Pulse requires a data source that is already published to Tableau Cloud. If you have existing workbooks, the underlying data sources are likely already published. To confirm, navigate to the Data Sources tab inside the Content section of your site.

If your data is in a flat file or a local database, you will need to connect to it in Tableau Desktop or Tableau Prep Builder first, then publish the data source to your site. The data source should include at least one date or datetime column and one numeric measure. Both are required to create a working metric definition.

Row-level security on the data source carries through to Pulse automatically. If a user does not have access to a row in the underlying data, that row will not appear in their metric digest.

If you are working primarily with CSV exports rather than a connected database, you may find it more practical to analyze those files directly in a tool like VSLZ, which generates charts and insights from a file upload without requiring a published data source or Tableau infrastructure.

Step 3: Create a Metric Definition

From the Tableau Cloud navigation, open Tableau Pulse. Click "New Metric Definition" to start the creation flow and connect to the published data source you prepared.

Fill in the following fields:

Name and description. The name appears in digests and search results. Use an unambiguous label: "Monthly Recurring Revenue" rather than "MRR" or "Revenue."

Values. Select the measure you want to track and choose an aggregation method: sum, count, average, or count of distinct values. For financial metrics, sum is usually correct. For headcount, count of distinct values is more accurate than a plain count.

Time. Select the date field that defines the metric's time series. Pulse uses this column to calculate period-over-period changes and detect trends. If your data source has multiple date columns, choose the one that represents the event timestamp most relevant to the metric.

Filters (optional). Apply default filters to scope the metric at creation time. For example, if you are creating a revenue metric for a single region, apply a regional filter here so that followers automatically see region-specific data.

Adjustable filters (optional). Mark specific filter dimensions as adjustable if you want each follower to personalize their view. If you publish a "Support Tickets Opened" metric and mark "Product Line" as adjustable, each team can follow the metric filtered to their own product without needing a separate definition.

Save the definition. It is now visible and followable by any user who has Connect and View permissions on the underlying data source.

Step 4: Follow Metrics and Configure Digests

Once a metric definition exists, users can follow it. From the Tableau Pulse home page, click "Browse Metrics" to see all available definitions. Open a metric and click "Follow."

Followed metrics appear in the Following section of the home page. Users receive a daily digest email showing a summary of each followed metric, including the current value, trend direction, period-over-period change, and any AI-generated insight if Einstein generative AI is configured on the site.

Digest timing is set at the site level by administrators. When following more than five or six metrics, users can mark selected metrics as Favorites and configure their digest to show only those, reducing noise. Off-cycle alerts fire when a threshold is breached on a non-digest day. To configure a threshold, open the metric definition and set a goal value or an alert range. For example, an alert when daily active users fall below a defined floor will fire immediately, regardless of the next scheduled digest.

Step 5: Connect Slack

If your organization uses Slack, administrators can configure Tableau Pulse to send digests to Slack channels or direct messages alongside or instead of email.

In Tableau Cloud Settings, find the Slack integration section. Authorizing the Tableau app in your Slack workspace requires admin access to both platforms. Once authorized, each user can choose Slack as a digest delivery channel from their personal notification preferences.

Each Slack digest includes the metric name, current value, and a brief AI-written summary of what changed since the last reporting period.

Reading AI-Generated Insights

Tableau Pulse surfaces several types of AI analysis inside each metric view:

Trend detection shows whether the metric is trending up, down, or flat over the selected period, with a confidence signal attached.

Contributor analysis identifies which dimension value is driving a change. If revenue declined, Pulse may flag that the drop is concentrated in a specific product category or region.

Outlier detection highlights specific dates or segments that fall outside the expected range based on historical patterns.

These insights appear in the digest email and in the metric detail view inside Tableau Cloud. According to Tableau's documentation, the Einstein Trust Layer ensures that actual data values are masked before being sent to the underlying language model, so sensitive records do not leave the organization in plain text.

The January 2026 Tableau release also added digest-disable controls at the site level, a signal that organizations with strict data governance requirements are increasingly adopting Pulse at scale.

Common Mistakes to Avoid

Using a non-published data source. Pulse cannot read local files or desktop extracts. If a metric definition loads no data, verify that the data source is published to the site and not saved only to a personal space.

Skipping the date field. A metric definition without a date column cannot generate trend analysis. Period-over-period comparisons (week over week, month over month) require a properly formatted datetime column in the source.

Mismatched permissions. Users who can see a metric definition in Pulse but lack View access to the underlying data source will see an empty metric. Set data source permissions before sharing definitions broadly.

Creating too many definitions at once. Start with three to five metrics that have a clear owner and a known audience. Definitions without active followers add noise to the browseable catalog and make it harder for new users to find what matters.

Summary

Setting up Tableau Pulse takes about 30 minutes for initial configuration, provided at least one published data source already exists on your Tableau Cloud site. The highest-value step is creating well-scoped metric definitions with clear names, correct aggregation methods, and relevant adjustable filters. Once those are in place, Pulse runs automatically and pushes AI-summarized metric updates to users on a schedule, removing the manual step of checking a dashboard each morning.