Tableau Pulse set up and first steps

Tableau Pulse and Tableau Cloud 2024.1 are here! How can you set it up in your Tableau Cloud site and start using it?

If you still don’t know what Tableau Pulse is, I’d recommend you to read first this post I wrote a few weeks ago explaining what it is, and how it helps making business decisions based on data faster and easier: What is Tableau Pulse?

How to set up Tableau Pulse

#

So, how can we activate Tableau Pulse? Easy. A Tableau Cloud Administrator can turn it on through the Settings page of Tableau Cloud and, in the General section, one of the first options available should say Tableau Pulse Deployment. Just check the Turn on Tableau Pulse box and decide if you want to make it available for all your Tableau users or just a concrete group. Then we need to enable the availability of Tableau AI so we can use generative AI to create the metric summaries and additional AI features, and finally save changes.

Once we have completed this step we should be able to see Tableau Pulse in the Tableau Cloud navigation bar on the left side.

Tableau Pulse first steps

#

Once we click Tableau Pulse another window opens and we should see something like this the first time after activation:

We should see a Following and Browse Metrics sections. Both empty as no metric has been created yet. Now we will need to create or first metrics based on our published data sources available in Tableau Cloud and subscribe ourselves to those metrics or, subscribe others. Lets see how to do all this.

Who can create a Tableau Pulse metric?

#

Any users with a Creator, Explorer (can publish) or Explorer Administrator role will be able to create what we call a metric definicion. The metadata definition of every metric.

How can we set up permissions on metrics?

#

There’s no specific permissions for metrics. Permissions are based on the underlying published data source the metric is connected to. This means a user will be able to see all the metrics if they can see and connect to the data source it’s connected to. If you don’t have permissions to see and connect to the data source, you will not be able to see and use the metric.

Is possible to have row-level security (RLS) in a Tableau Pulse metric?

#

Yes. Tableau Pulse metric respect the row-level security of the data source if it’s set up. This means that the use of Tableau’s Virtual Connections combined with Pulse metrics can be very useful and powerful to make sure every user sees and gets insights only of the data they should have access to.

What is the general workflow of Tableau Pulse?

#

It’s quite easy. A Creator, Explorer (can publish) or Explorer Administrator should create metric definitions based on published data sources they have access to (I’ll explain how to do this a bit later). Once the metric definition is created, any user with access to the data source will also have access to the metric definition and will be able to subscribe to the default metric. Additionally, users will be able to adjust the default metric to their needs by adjusting the filters, time period analysis or period comparison. Lets see an example:

• A Creator creates a metric called “Total revenue” that measures the revenue for the company based on the data source “Financial Data” and includes additional dimensions to the metric: Product, Country, Account Executive. Once the metric definition is ready, he or she includes all the Sales Managers as Followers of the metric.
• One of the Sales Managers opens Tableau Pulse and sees already the metric in his or her homepage. Opens the metric and adjust some filters to track only a concrete set of Products he/she is interested in and starts following that version of the metric or what we call related metric. Now two metrics will appear in the homepage: the default view of the Total revenue metric with all the products and another one with the Total revenue only for the products he/she is more interested in.

Metric definitions and related metrics

#

At this point, is important to clarify and understand the difference between metric definitions and related metrics.

When a Creator, Explorer (can publish) or Explorer Administrator creates a metric, what he or she is really creating is a metric definition. A metric definition contains metadata that forms the unique source of truth for all the related metrics to that metric definition. What does this metadata include? Information like:

• The metric name.
• The measure used and the aggregation (example: sum of sales).
• The date field to be used (example: Order date).
• Additional dimensions to filter and slice & dice the data (example: Product, Country and Account Executive).
• The measure number format (number, percentage, currency, etc.)
• How the AI refers the measure, both in singular and plural (example: euro and euros)
• The measures sentiment when it goes up (if the measure goes up, is this favourable, neutral or unfavourable?).
• The type of insights available for the measure (we can enable / disable each one individually).

This metadata will not change unless we edit the metri definition. This way, the metadata is shared for all the related metrics dependant of the metric definition.

Once we have the metric definition created, it has some default properties like the time period analysis and period comparison. If a user has access to the metric definition, they can adjust those default characteristics to their needs. When a user adjust those, a new related metric is created automatically and will be available for anyone that can access the metric definition. Those related metrics can be then followed by the users and will be available in each users main Tableau Pulse page.

Lets see an example. We want to measure and track our Sales in total, but also our Sales in a concrete country with Tableau Pulse. The metric definition will be the same in both cases, and I can subscribe to two related metrics, one with all countries and another one with a Country filter applied. From the default metric what I’ll be able to adjust is:

• The time analysis granularity: Current month until today, current quarter until today, etc.
• The comparison period: Compared to the previous period or the same period of the previous month.
• Filters applied: Based on the dimensions included in the metric definition.

Now that we know what a metric definition and related metrics are, lets see an example of how to create one.

How to create a Metric Definition in Tableau Pulse

#

To create a metric definition, a Creator, Explorer (can publish) or Explorer Administrator has to click on the New Metric Definition button from Tableau Pulse home page.

A new window will pop up showing the published data sources from Tableau Cloud we have access to:

Now we can search and select the data source we want to use and click Connect. We can start then defining and setting up our metric definition.

Setting up a Metric Definition in Tableau Pulse

#

At this point you should see two main sections in the top menu of the creating a metric definition window: Definition and insights. Lets start with the Definition. First thing we need to specify is the name for the metric and, even if it’s not a mandatory field, my recommendation is to make sure we fill in the description. A very detailed description. Have in mind that it will be key for end users to trust the metric they are following. Having a high detailed description will be key for that. After the description, we must select the Measure from the data source that we want to track in our metric.

Now we have the name, description and measure. But we need a few more things to set up. We need to specify the Time dimension, this is, the date field from our data source to track the metric across time, and the primary time comparison (vs prior period or vs prior year). As you can see in the image below, we could also click Create Advanced Definition to create a Metric Definition based on calculated fields if the measure we want to track is not in our data source but we can calculate it.

The last optons to set up include the Adjustable metric filters, where we should specify the additional dimentions to be available to slice and dice the Metric, filter it, and get additional insights. The last option in this section is the Number format, used to specify how we want to treat the number of our metric (number, percentage, currency) and also the units we want to use by the AI inisghts generated with natural language, both in singular and plural.

Setting up the Insights of a Metric Definition

#

Don’t save the metric definition yet. We can also configure additional parameters in the second tab of this menu: the Insights tab.

In this section we can configure two things. First, the sentiment when the metric goes up: it’s favorable, netural or unfavorable? And second, we can enable or disable the different type of inisghts that Tableau Pulse will analyze and generate based on the current status of the metric.

Now we are done and we can click Save Definition. Then, the default related metric for this metric definition will be available. We can adjust the dimensions, period comparison and filters. And follow the metric if we want clicking the + Follow button.

Each metric will provide automatic insights, the option to breadkdown the current metric value based on the selected time-period and the dimensions added and also natural language explanations of the current situation.

If we go back to the Tableau Pulse home screen by clicking the Back button or the Tableau logo on the top-left of the screen, we will see a summary on the top of the screen generated with generative AI with an overview of the metrics, and also we will see all the metrics we are following. Each of them it’s clickable and available to get AI insights and get more detailed information to help us understand what’s going on and make better business decisions based on data.

How to see all the Metric Definitions created

#

How can we see all the metric definitions that have been created and that I have available? We can navigate to the Browse Metrics section. There we will see a full list of all the metric definitions created by ourselves or other users and that we have access to.

How to subscribe or follow to a metric

#

And if I want to subscribe to a metric? Easy too. Just select the metric definition you are interested in from the Browse Metrics menu. In the next screen you will see the details (the metadata) of the metric definition and also all the related metrics already available.

this means I can just go to this section to check the filters and views other users have created based on each metric definition, check them, review them and follow them directly. In the image above we can see that related to the metric Horas de sueño diarias (sorry for the spanish, this is the amount of hours I sleep every day), there are two related metrics already available: one that I follow with a quarterly view of the metric and another that I don’t follow with a monthly view.

Customizing a metric

#

In case I need to edit a Metric Defininiton, I could click Edit Definition in this same screen if I have the permission to edit de metric definition and change or update it. Another way to modify a metric definition is from the list of metrics in the Browse metrics menu, clicking in the actions button and selecting Edit.

In Tableau Pulse hompage we can also check the additional options available for a metric we are following. Where I can unfollow the metric, see the metric details or manage followers, in case I want to add users or groups of users as followers.

The user preferences

#

Last but not least, we can edit our user preferences for Tableau Pulse by clicking the user button on the top-right of the screen. Nowadays there’s not a huge amount of options there, but we can configure if we want to receive a digest of our metric status in daily, weekly or monthly basis, and if we want to receive it by email, Slack or both.

I hope this fast overview of how to start using Tableau Pulse helps you getting the most of it since the first minute!