Skip to main content
Gainsight Inc.

Create Datasets in Data Designer

Overview

Data Designer helps admins build Datasets from disparate data sources. Using these datasets, you can explore and analyze the data to see if it’s valuable or not, and then make it permanent across Gainsight. This article will walk you through the configuration process of how to Prepare, Analyze, and Configure (schedule) a Dataset in Data Designer.

Before you get started with the Dataset creation process, Gainsight recommends you to refer to Data Designer Overview article.

In this configuration article, we used the below mentioned use case to demonstrate the process of creating a dataset. 

Use Case

For instance, you might want to know the customers who are raising support tickets are actually using the product correctly or not, and you might also want to know if those customers have good health scores when compared to other customers who are not raising support tickets. Assume that you have Case Data (Support Data) in a Salesforce Object, and you have other data sources in multiple Gainsight objects like, Usage and Company. With Data Designer, you can merge and transform the data from these data sources, to correlate and derive real and powerful insights.

Limitations

  • The Correlation formula in a Dataset Preview doesn’t work the same as the Correlation formula on the Analyse Preview. The numbers might slightly differ. The Correlation formula in Dataset Preview works precisely.
  • The Median value you see in the Analyze page is approximate, and therefore the value you see in Analyse preview is different from Data Preparation preview.
  • Assume, you have built a report on the Dataspace 2.0 and added it to the Dashboard. And later, if you remove any fields from the Dataset and run it, the report in the dashboard stops working and throws an error.
  • The Datetime data in the Preparation Preview is shown in Org Timezone, the Datetime data in the Analysis page is shown in the User Timezone, and the Datetime data in the Execution Logs is shown in the UTC Timezone.
  • In a Dataset, if you have included any picklist or multi-picklists data type fields from Salesforce. In Filters, these picklists data type fields are converted to String data type fields, after the data preparation.
  • WHOID and WHATID type fields in a dataset display IDs.
  • Assets of the Data Designer cannot be migrated using X-org Migration Tool.

Create New Design

To create a new design:

  1. Navigate to Administration > Analytics > Data Designer . You can see the existing list of designs, if required you can edit any of the designs you wish to. For information on Designs Listing page and Options, refer to Designs Listing Page and Options article.

  2. Click New Design. The Create Recipe page appears, where you can see four tabs, namely:

    1. Details
    2. Preparation
    3. Analysis
    4. Configure

Flow.gif

Details

Enter the following information in the Details tab. In this use case, the following information is used:

  1. Design Name: Case, Company and Usage Data.
  2. Select Folder: Select the folder in which you want to add this design, if you have already created a folder in the Designs Listing page (or) By default, the design will be saved in the Uncategorized folder.
  3. Description: [Optional] Merge Case, Company and Usage Data to derive real insights.
  4. Click Prepare on the bottom right corner of the screen, to save the design and you will be navigated to the Preparation tab.

Details.gif

Preparation of Dataset

Data Designer enables admins to prepare data with confidence. For instance, if you are looking for insights like Correlation in the prepared data, and if you think this insight is not available in the prepared data, you can discard the dataset and create a new one.

In the preparation step, you can merge two different objects and create a Dataset. If required, you can Transform a dataset to apply Filters, Formulas, Case Expressions, etc. 

For information on the Functionality Details in Preparation tab, refer to Preparation Details in Data Designer article.

Create Dataset 1 (Case Data)

To create first dataset:

  1. Select the required source from the Data Source dropdown. Once you select the Data Source, you can see all the objects under the selected Data Source. Here, the data source is SFDC_Connection.
    Important: If the data you want to analyze is not from Gainsight, you can make use of custom objects in Gainsight to load your data from external sources or if your data is available in Amazon S3 buckets (Gainsight managed or Custom buckets) you can create datasets using this data. 
    For more information on how to use data from S3 in Data Designer, refer to Use S3 Data in Data Designer article.
  2. Drag and drop the required Object from the Objects list to the Canvas screen. Here, the object is Case_Demo.
  3. Select the required fields you want to add to the dataset. In this use case, select the following fields:
  • CaseID
  • Company Name
  • Priority
  • Case Subject
  1. Click Select.
  2. Click Save.

First Dataset fields.gif

Hover to the dataset created, and click the Preview/Eye icon to view the data. In Preview, you can view the sample data and might not contain the entire data.

Preview.gif

Click Refresh on the Preview tab to refresh the sample data, if you have added/deleted the fields etc.

Refresh.gif

Create Dataset 2 (Company Data)

To create second dataset:

  1. Select the required source from the Data Source dropdown. Here, the data source is Matrix Data.
  2. Drag and drop the required Object from the Object list to the Canvas screen. Here, the object is Company_Demo.
  3. Select the required fields you want to add to the dataset. In this use case, select the following fields:
  • ARR
  • Name
  • Real Score
  • True NPS
  • Score Label
  1. Click Select.
  2. Click Save.

Second Dataset.gif

Note: For Numeric type fields, you can change the number of decimal places by clicking the Settings/Gear icon in the Fields tab. 

Merge 1st and 2nd Datasets

To merge the first and second datasets, to correlate the Case and Company data:

  1. Click Options on the dataset you want to merge, and then click Merge and select the required dataset. Dataset Details window appears.
    Note: You can view all the datasets available in the selected design.
  2. [Optional] Rename the Dataset Name as required. Here, the dataset is renamed as Case and Company Data.
  3. Select the required join condition in the Join tab. Here, for this use case, select the Inner join condition. For more information on Joins, refer to Joins section in Preparation Details article.
  4. Select a field from each dataset to set the criteria. For instance, the Company Name in Company_Demo dataset is Name, and Company Name in Case_Demo dataset is Company Name.
  5. Click + to add multiple join conditions, if required. This helps filter the records based on your business requirements.
  6. In the Fields tab, you can select/deselect the fields as required. For this use case, deselect the Name field to avoid multiple records, as you have Name and Company Name fields from both datasets.
  7. Click Save.

Merge first.gif

  1. Click Preview, and see if the data is valuable or not. If you think that the data is not valuable, you can edit the dataset.

Merge Preview.gif

Create Dataset 3 (Usage Data)

To create the third dataset:

  1. Select a source from the Data Source dropdown. Here, the data source is Matrix Data.
  2. Drag and drop the required Object from the Object list to the Canvas screen. Here, the object is Usage_Demo.
  3. Select the required fields you want to add to the dataset. In this use case, select the following fields:
    • Name
    • Page Views
  4. Click Select.
  5. Click Save.

Transform Dataset 3

To transform the third dataset:

  1. Click Options on the dataset (Usage_Demo), and click Transform. Dataset Details window appears.
  2. In the Fields tab deselect the Date field to avoid the multiple records in the same month.
  3. Select Group By for the Company Name field, and set the Aggregation of the Page Views field to Sum, to achieve Sum of Page Views by Company Name.
    Note: In the Transformation task, you can also apply Formula Fields and Case Fields on the fields added to the dataset. For information on how to create Formula Fields and Case Fields, refer to Formula Fields and Case Fields section in Preparation details article.
  4. Click Save.

Third Dataset.gif

Merge ‘Case & Company’ and Transformed ‘Usage’ Datasets

Perform Merge on Case, Company and Transformed Usage datasets to create an output dataset.

  1. Drag and drop Case and Company dataset over the Transformed Usage dataset to merge.
  2. [Optional] Rename the Dataset Name as required. Here, the dataset is renamed as Case Company and Usage
  3. Select the required join condition in the Join tab. Here, select the Inner join condition.
  4. Select a field from each dataset to set the criteria.
  5. Click + to add multiple join conditions, if required. This helps filter the records based on your business requirements.
  6. In the Fields tab, you can select/deselect the fields as required. Here, deselect the Name field to avoid multiple records, as you have Name and Company Name fields from both datasets.
  7. Click Save.

MErge three.gif

  1. Click Preview to view the sample merge data of the datasets.

Final Preview.gif

Analyse Dataset

In the Analysis tab, you can report and analyze the output dataset created during preparation, and see if the data is valuable or not. During your analysis, if required, you can apply Filters, Formulas, modify aggregations, change visualizations, etc. For more information on Functionalities in the Analysis tab, refer to Analysis Details in Data Designer article.

To Analyse the output Dataset:

  1. Click Analyse on the bottom right corner of the preparation screen.
  2. Once you navigate to the Analysis tab, you can see the following image. Click Run Now to associate a dataset with the design, and you can see that the Dataset is being prepared for analysis.

Note: Dataset Preparation for Analysis may take several minutes, if required you can navigate to the Designs List page or Configure page by clicking List and Configure respectively.
 

Run and Refresh.gif

Once the Refresh is completed, you will receive an execution status email. And, the dataset is now ready for your analysis and you can see the fields from the output dataset under Fields section.

Analysis main.gif

Note: If you modify the already saved datasets, and then navigate to Analysis tab, you can see a message at the top of the Analysis tab that you can either Refresh Dataset for the latest data or Dismiss it to continue analysing the existing dataset.

Modify Refresh.gif

Build Reports to Analyze Data

To explore and analyze the data, you can build reports in the Analysis tab. 

Create Report 1: To see if the customers who are raising the tickets are actually using the product or not. 

  1. Drag and drop the Case ID and Page Views fields to the Show Me section. 
  2. Drag and drop the Company Name field to the Group By section.
  3. Click Analyse. You can now see the records being displayed in a tabular report.
  4. Hover to the right of the screen and select the required visualization type.
  5. Click Save Report. Save Report window appears.
  6. Enter the Report Name in the Report Name textbox.
  7. Click SAVE. Click the Hamburger button to see all the saved reports.

Report 1.gif

Create Report 2: To see the customers who are raising the tickets are Healthy or not.

  1. Drag and drop the Case ID field to the Show Me section. 
  2. Drag and drop the Score Label field to the Group By section.
  3. Click Analyse. You can now see the records being displayed in a tabular report.
  4. Hover to the right of the screen and select the required visualization type.

Report 2.gif

  1. Click Save Report. Save Report window appears.
  2. Enter the Report Name in the Report Name textbox.
  3. Click SAVE.

Report 3.gif

You can also publish these reports to Report Builder repository. For more information on how to publish reports, refer to Analysis Details in Data Designer article.

Configure

After your analysis, if you think that the dataset has some real value, then you can make it permanent and universal. Once a dataset is permanent, it is saved as a Dataspace under MDA, and you can use in other functionalities of Gainsight like, Reporting, Rules, and Journey Orchestrator (JO).

To configure the Dataset:

  1. Navigate to the Configure tab (or) click Configure on the bottom right corner of the Analysis screen.
    Note: The dataset you prepared is temporary and will expire in 15 days. You can see the notification message on the top of the Configure tab.
  2. Enable the Dataspace toggle to make the dataset permanent and universal.
    Note: Once the Design is permanent, it is called a Dataspace and is available across the Gainsight and you can view the schedule options to refresh the dataspace.
  3. Schedule: Enable Do you want to refresh this Dataspace at regular interval? toggle button

  4. From the Schedule Type dropdown list, select either Basic or Advance, based on your requirement.

  5. Basic Schedule: You can schedule the refresh either Daily or Weekly or Monthly.

Daily:  You can refresh the data space either Every Weekday (Mon - Fri) or Everyday

Weekly: You can select the preferred day in a week to refresh the data space on a weekly basis.

Monthly: You can either select a preferable Date or Day in a month to schedule a refresh on monthly basis.

To schedule a Refresh:

  1. Select either Daily or Weekly or Monthly. 
  2. Select the Start Date and End Date.
  3. Set the time in the Select Time text box.
  4. Set the required Time Zone from the Select Time Zone dropdown list.
  1. Advance Schedule: Advanced Scheduler in Data Designer uses Cron Expressions to refresh the Data Space. This enables you to refresh the data more frequently, compared to the Basic scheduler. And, Advanced Scheduler also addresses some of the advanced use cases including refreshing the Data Spaces as frequently as every two hours, or on a less frequent basis to match business processes, such as on the 1st and 15th of the month.

    Cron Expression: This is a string comprising 6 or 7 fields that are separated by spaces and defines the details of a schedule. Each field represents a sub-expression and allows only specific values. These values combined with special characters define the schedule.

    Note: The minimum time period between data space runs is two hours.

    1. Input the cron expression in the Enter Cron Expression text box. For information on how to input the cron expression, refer to the Cron Expression for Advanced Scheduler in Data Designer article.

    2. Select the Start Date and End Date.

    3. Set the required Time Zone from the Select Time Zone dropdown list.

  2. Notification Preferences: Once you enable the Notification Preferences toggle, the logged-in user receives an Email Notification of the Execution Summary of that particular Data Design on every run (as configured in Schedule section). If required, you can also include additional email addresses as mentioned below:

    1. Enter the email addresses in the Mention Emails For Failure textbox, to send failure notifications to those emails.

    2. Enter the email addresses in the Mention Emails For Success textbox, to send success notifications to those emails.

  3. Click Done.

Now, navigate to the Context Menu of the Dataset, and click Run Now to run the dataspace for the first time, and later on the next runs happen automatically as per the defined schedule.

Consumption of Dataspace

You can consume the Data Space in Report Builder, to create reports and then add it to the GS Home Dashboards.

To create Reports from Report Builder:

  1. Navigate to Administration > Analytics > Report Builder.
  2. Select the Data Space from the MDA Data Source.
  3. Add the required fields to Show Me and By.
  4. Click RUN, and Save the report.

Report Builder.gif

You can now add the saved report to a dashboard, for your CSMs to view on GS Home.

ezgif.com-crop.gif

Additional Resources

  • Was this article helpful?