Use Admin Insights to Create Custom Views

You can get more visibility into your Tableau Online deployment by using Admin Insights.

Admin Insights is a Tableau Online-only project that is pre-populated with carefully curated data sources and a pre-built workbook of your site's data. Using the resources available to you through the Admin Insights project, you can create custom views to help answer a range of common questions you might have about your site.

For example:

  • What's my Tableau Online adoption rate in my organization?

  • What are common trends around the site’s deployment?

  • What content is popular?

  • What are my users doing?

  • How should licenses be allocated?

Get started with Admin Insights

To learn more about enabling Admin Insights for your site, see Enable and Manage Admin Insights.

Connect to Admin Insights data

If you're a site admin or someone who has been granted access to the Admin Insights project, you can access the Admin Insights data sources directly from Tableau Online using Web Authoring or through Tableau Desktop.

From Tableau Online

  1. Sign in to Tableau Online and navigate to the Admin Insights project.

  2. Select Create > Workbook and select one of the data sources to get started.

From Tableau Desktop

  1. Open Tableau Desktop, under Connect, select Tableau Server.

    Note: If you're not already signed in to Tableau Online, in the Tableau Server Sign In dialog box, click the Tableau Online hyperlink. Enter your Tableau Online credentials and then click Sign In.

  2. In the search box, type "TS." A list of Admin Insights data sources displays.

  3. Select a data source and click the sheet tab to get started with your analysis.

Start creating custom views

As you think about how you want to approach the analysis of your site and user metrics, consider some of the following questions that organizations commonly ask:

  • What are the most popular views or data sources?
  • Who are the most active users (i.e., who are the power users)?
  • What are the most common tasks performed on the site?
  • How many licenses are being used?
  • Who hasn't signed in to the site in more than 90 days?

Explore the pre-built workbook

To help you answer the above questions (and more), go to the pre-built workbook, Admin Insights Starter, to see what kind of insights it can offer. The pre-built workbook is intended to serve as template on which to build more detailed dashboards and workbooks that go more in depth and address questions you have that are unique to your site deployment or organization.

Focus: Site activity
Data source: TS Users

Use this dashboard to explore high-level site activity by user role.

This dashboard shows you just some of the ways you can analyze your site’s log and activity metrics. Although Admin Insights captures 90 days worth of data, the “Last Publish” and “Last Access” dates can go back to as early as the date the site was created.

Do more with this dashboard:

  • Change the "Inactivity threshold" (yellow by default) using the parameter control in the upper-right corner.
  • Create URL actions that can email users based on their activity type. For example, email users when they haven’t signed in to the site, accessed content, or published content in the last 90 days.
  • Create additional views for this dashboard that compares the last 90 days (default) of activity (sign in, access, or publish) to the last 30 days, 60 days, and all days.

Focus: Sign-in activity
Connects to: TS Events (primary), TS Users (secondary)

Use this dashboard to explore trends in sign-in activity.

Using the “Select Events or Distinct Actors” selector on the left, you can see your site’s weekly, daily, and hourly sign-in-related activity. Because there are two data sources behind this dashboard, you can see sign ins by all users, not just the users that have sign-in activity in the last 90 days.

Do more with this dashboard: Duplicate the dashboard and keep only the “Login user breakdown.” Then change the filter on the “Last Login Date” to null to see a list of users who have never signed in to the site. You can also email them depending on what kind of action you want users to take.

Focus: Content access and usage
Connects to: TS Events

The information in the dashboards show you the following type of information:

  • View access activity: When a view has been accessed and by whom.
  • Data source access activity: When 1) a published data source has been connected to through Web Authoring or Tableau Desktop or 2) a user has viewed or published a workbook that uses the published data source.
  • Workbook publish activity: When a workbook has been published and by whom.
  • Data source publish activity: When a published data source has been published and by whom.

Do more with this dashboard: Using the “Project Name” selector in the upper-right corner, select a project on your site to filter on. Note: This filter affects all sheets in the Admin Insights Starter.

Focus: View load times
Connects to: TS Web Requests

Use this dashboard to better understand the load times for views. Here, load times mean bootstrap sessions events. These events are registered any time a request to load a view is received and then loaded by the server. The bootstrap session time does not include the time it takes for the browser to send the request to the server and render the view.

Do more with this dashboard: Using the “Project Name” selector in the upper-right corner, select a project on your site to filter on. Note: This filter affects all sheets in the Admin Insights Starter. Using the "Data Range" selector in conjunction with clicking on a date in the "Median Load Times by Date" view, you can explore deeper to find points of interest.

Request access and share feedback: This dashboard is available as an in-release beta feature by request. To join the beta or provide thoughts and ideas about the dashboard, sign in to the Early Feedback site.

Explore the data sources

Alternatively, you can connect to the Admin Insights data directly. Explore the data on your own by hovering over each field (both dimensions and measures) to read a description of the data that's being captured.

TS Events is an extract data source that functions as a master audit data source. It contains data about the various events happening on your site, including sign-ins, publishes, and accessed views

TS Users is an extract data source that contains data about your users such as remaining licenses, site roles, and workbooks and views owned by a user.

TS Web Requests is an extract data source that contains data about the view load times to help those users generating content better understand the experience of the users who are consuming the content.

Request access and share feedback: This data source is available as an in-release beta by request. To join the beta or provide thoughts and ideas about the data source, sign in to the Early Feedback site.

Example custom views

You can follow the example procedures below to start creating your own custom views.

Example 1: What are the most popular views?

  1. Connect to the TS Events data source using one of the procedures listed above in Connect to Admin Insights data.

  2. From the Data pane, drag "Item Name" field to the Rows shelf and "Number of Events" field to the Columns shelf.

  3. From the Data pane, drag "Field Type" to the Filters shelf, remove the selection next to the All check box, and select the Data Source check box.

Example 2: How many licenses are being used?

  1. Connect to the TS Users data source using one of the procedures listed above in Connect to Admin Insights data.

  2. From the Data pane, drag "Measure Names" to the Rows shelf and "Measure Values" to the Columns shelf.

  3. Right-click the "Measure Names" field in the Rows shelf and select Show Filter.

  4. Click the Measure Names filter drop-down menu, and select Customize > Show Apply Button.

  5. In the filter, select "Total Allowed Licenses" and "Total Occupied Licenses" check boxes, and then click the Apply button.

 

Thanks for your feedback! There was an error submitting your feedback. Try again or send us a message.