Using Multiple Datasets in a Dashboard (CX)

About Using Multiple Datasets in a Dashboard

As your experience research programs grow and become more complex, you may already have individual datasets you’ve made that would take time to rebuild into one big dataset. You may also just want to visualize different datasets in one dashboard. You can resolve both of these goals by adding multiple datasets to a single dashboard.

By adding multiple datasets, you can have widgets from different datasets presented side by side, or even show multiple datasets’ results in the same widget. You can even use linked filters that will work regardless of the dataset, creating a seamless experience for your dashboard viewers.

Example of Multiple Datasets in a Dashboard

Let’s say you want to show all the distribution channels where you’ve collected data, such as chats, emails, phonecalls, and so on. You want to show a breakdown of responses across channels and allow viewers to filter interactions by date, regardless of where the data came from. But you have distribution channels in two fairly different datasets: one imported from Discover, and the other from Qualtrics.

1. You need to first link the date fields from the two different datasets.
2. Then, add a page level filter for date.
3. Lastly, create two widgets, one for each dataset, each broken out by distribution channel.

As the date filter is adjusted, both widgets are affected. This allows dashboard viewers to see a wide range of data from different sources without you, as the dashboard builder, having to create complex joins or build a new, combined dataset.

Instead of creating two separate widgets, you can also create a single table, line, or bar widget that shows data from multiple datasets.

Example: Both of the bar widgets below show data from both our Qualtrics and Discover datasets. The left widget shows a simple response count for each dataset. The right widget shows response count for each dataset broken out by distribution channel, which is a linked field.

Types of Dashboards that Support Multiple Datasets

Multiple datasets are only available in Dashboard projects (CX Dashboards).

Multiple datasets are not available in Survey Results Dashboards, BX Dashboards, or in any Employee Experience dashboards.

Linking Fields Between Datasets

Before you can use multiple datasets to build reports for your stakeholders, you need to add another dataset to your dashboard, and link fields that should be the same across datasets.

There is no limit to the number of linked fields.

1. Go to the dashboard settings.
   
2. Go to Dashboard Data.
   
3. Click Add dataset.
4. Read the warning about multiple dataset limitations. If you want to continue, click Confirm.
   
5. Select an available dataset, or create a new dataset to select.
   
6. Click Add.
7. Click Manage linked fields.
   
8. If you need to, use the search or the display to find the specific field you want to link.
   
   Qtip: The Display filter lets you narrow down results to all fields, just unlinked fields, or just linked fields.
9. You’ll see each dataset as a column. The default dataset will be the leftmost column.
   
10. Each row is a field from the default dataset that you can map to a field from your second dataset.
    
11. Select a dataset field to match it to another field on the left.
    
    Example: The default dataset has a field called “Survey Metadata – Recorded Date.” We mapped this to the “Document Date” field in our second dataset.
12. Click Confirm.

Auto-link fields

We provide a tool that makes it easy to link fields that have similar names. By using auto-link, you can even link multiple sets of fields at a time.

1. Click Auto-link fields.
   
2. Review the changes.
   
   Qtip: Only auto-linked sets of fields will be displayed.
3. If you want to link these sets of fields, click Confirm.

Custom Linked Fields

Custom linked fields are useful if you have 3 or more datasets in your dashboard. The purpose of a custom linked field is to let you link fields from datasets without including anything from the default dataset (i.e., the primary dataset).

1. Have 3 or more datasets added to your dashboard.
   
   Qtip: Click Add dataset to add more datasets to the dashboard.
2. Click Manage linked fields.
3. Click Add custom linked field.
   
4. Select the fields that should be linked across datasets.
5. Click Confirm.

Clearing Linked Fields

You can easily disconnect linked fields in a few different ways:

1. Clear an entire column of linked fields: Remove all of the linked fields from a dataset. Click the 3 dots next to the dataset name and select Clear column linking.
2. Unlink a specific mapping: Click the menu and select Not linked.
3. Clear an entire row of linked fields: This option can be especially helpful for unlinking a field from 3 or more datasets. Select the 3 dots next to the field, then select Clear row linking.

Managing Datasets in a Dashboard

In Dashboard data, you can manage all of the datasets you’ve added to your dashboard. Click the 3 dots to see more options:

• Edit dataset: Edit the individual dataset. This is where you go to edit fields, add new fields, change recodes, and more. For more details, see data mapper and data model
• Lineage and usage: This option opens the Data page in a new tab and shows you all the places where the dataset is being used. See more on Lineage and Usage.
• Remove: Remove the dataset from the dashboard.
  Qtip: You cannot remove the default dataset from the dashboard.
  Warning: Make sure you consider linked fields, filters, and widgets before you delete a dataset from your dashboard!

Creating Linked Filters

Once you’ve linked a field across datasets, the next step is to create a linked filter. Let’s say you want to display data from multiple datasets in your dashboard. A filter that only references one dataset can only affect widgets from the same dataset. But a linked filter could narrow down data in all of the widgets, no matter the dataset.

1. Click Add Filter.
   
2. Choose between your datasets.
3. Select a field to filter by. Fields present in multiple datasets (linked fields) have a link icon.
4. Decide whether to select Only show values from the referenced dataset.
   
   • Selected: Only let dashboard viewers use the options available in the dataset you chose in step 2.
     Example: Your selected dataset has the Distribution Channels of Chat, Email, Phone, and Reviews. Regardless of the Distribution Channels available in your second dataset, you’ll only see these 4 options listed.
   • Deselected: Dashboard viewers can use the options available in either dataset.
     Qtip: The option Only show values from the referenced dataset only appears when filtering a linked field by certain field types. When not selected, all values from both datasets’ fields will appear in the filter value selection.
     Qtip: This option exists because the values can vary for a linked field across datasets. The more you align recode values between the separate datasets, the more seamless the filter experience will be.
5. Otherwise, your customization options are similar to other dashboard filters.

Using Linked Filters

Filters in the dashboard will be separated based on the datasets they can impact. Linked filters can impact multiple datasets.

Linked filters behave a lot like regular, unlinked filters. For example, if you filter by a value, and your linked widget doesn’t have data for that value, it won’t show data.

Only widgets that reference the datasets in the linked field will be affected by the relevant linked-field filter.

Example: You have 3 datasets. In your third dataset, you don’t link the date field. When you filter by date, the third dataset’s widgets do not change. (In the screenshots below, we show how widgets react to a changing date filter, with the third widget unaffected.)

Saving Linked Filters

You can save filters so you can go back to your favorite configurations later. When you have multiple datasets in your dashboard and / or linked filters, this can cause saving filters to behave a little differently.

When you open the filter menu, you’ll see that there’s a section for each dataset. Thus, filters are saved, selected, and managed separately by dataset, even if they have linked fields. This also means you can save a linked filter under each dataset it belongs to, thus saving it multiple times.

Example: For example, here our “Distribution Channel” field is linked between our Qualtrics and Discover OXM datasets, so we’ll save the filter 2 times, once under each dataset. Later, to reapply this filter, I could use the version saved under either dataset.

It’s important to keep in mind that when you save a filter multiple times under separate datasets, these are treated as separate filters. That means new changes can only be saved to whichever filter you currently have selected.

Adjusting multiple dataset filters and saving

You can have filters from linked fields and different datasets add to the same page at the same time.

When you adjust your filters and save those configurations for later, it doesn’t matter what datasets your fields come from. When you save the filter, it saves all of the configurations.

Adjusting a Widget’s Referenced Dataset

Once you’ve added multiple datasets to your dashboard, you can choose which dataset is referenced in each widget you create. This can be thought of as the widget’s primary, or default, dataset.

When you edit a widget, you can adjust the Referenced dataset being used.

Widgets that let you reference different datasets:

• Breakdown bar
• Line chart
• Horizontal bar chart
• Vertical bar chart
• Donut / pie
• Gauge chart
• Star rating
• Scatterplot
• Table
• Record table
• Digital Opportunities
• Response ticker

Widgets that only reference the default dataset:

• Breakdown trends
• Text iQ bubble chart
• Journey chart
• Multiple source table
• Breakdown table
• Text iQ table
• Key drivers
• Focus areas
• Coaching priorities
• Legacy / deprecated widgets

Adding Multiple Datasets to the Same Widget

A few widgets allow you to show data from separate datasets at once. This is done by adding multiple metrics and changing the dataset your metrics reference. From there, you can even break out the data by linked fields, giving you dynamic options for side-by-side dataset comparisons in the same widget.

The only widgets that can show multiple datasets at the same time are line, bar, and table widgets.

1. Add metrics to your widget.
2. Click the metric.
3. Adjust the Dataset selected.
   
4. Once you add multiple datasets, breakout settings (e.g., Data series in bar charts, Columns in tables) will only let you use linked fields.
   

Multiple Datasets in Other Dashboard Settings

In addition to fields, datasets have other features you can customize in a dashboard, such as their custom metrics, benchmarks, and more. When you go to edit these dashboard settings, you can decide which referenced dataset you want to work on. That means these features don’t allow you to create items that are linked between datasets. For example, you can make custom metrics within the same dataset, but not across different datasets.

Dashboard settings where you can navigate between multiple, separate datasets:

• Text iQ
• Custom metrics
• Benchmarks
• Response weighting
• Translations

Other settings, like theme, workflows, date & time, and action plans, are set for on the entire dashboard, rather than just separately for each dataset.

Controlling Dataset Access

User access is controlled on a dashboard by dashboard basis, not on a dataset by dataset basis. That means anyone with editing access to a dashboard can also access and edit any datasets inside. You can control access to the dashboard and its datasets using roles.

When you restrict data by dashboard fields, you must choose a dataset first. From there, you can choose fields, linked or not, to build data restrictions. If you have multiple datasets in your dashboard, and want to use data restrictions, we recommend making sure those data restrictions use linked fields.

FAQs