Once you have access to Journey Insights Hub and a workspace, you can create a text analysis project by uploading a prepared dataset. Your file should include the open-ended text you want to analyze and may also include metadata such as dates, scores, categories, or other details that provide context.
Metadata can later be used to filter and analyze specific subsets of records. Numerical and score fields may also be used in Drivers analysis.
Only users with the Admin or Creator role can create a new text analysis project.
Start a new text analysis project
From the Projects page, select + New project.
You can return to the Projects page by selecting the Journey Insights Hub name or logo in the upper-left corner of the screen.
On the text analysis project creation page, you can add your dataset in either of two ways:
Select Choose file and locate the file on your computer.
Drag and drop the file into the upload area.
Only one file can be uploaded at a time. Once the file is selected, Journey Insights Hub displays a preview so you can review the data before creating the text analysis project.
Text analysis project information
The Project Info section contains the basic settings for your new text analysis project.
Text analysis project name – defaults to the name of the uploaded file and can be edited
Description (optional) – can be used for information about the text analysis project
Workspace – defaults to whatever Workspace you were using when you clicked New project. If you have multiple Workspaces, you can select a different one before creating the text analysis project
Language – select the language used in your dataset out of the 14 supported languages. For multilingual data, separate the records by language and create a different text analysis project for each language
Review and assign your data fields
The Data Summary panel appears on the left side of the page. It contains one item for each unique column header in your file.
Use the menu on each item to confirm or change its assigned data type. Journey Insights Hub analyzes one Text field. If your file contains multiple open-ended text columns, assign each at Text and use Configure Text Columns to combine the selected columns into a single analysis and add a Question metadata field identifying the source column.
Other columns can be assigned as metadata. When multiple columns use the same header, they appear as one grouped item in the Data Summary panel.
How data types are assigned
Journey Insights Hub uses your column headers to determine the initial data type when they follow the supported naming format.
For example:
Feedbackis assigned as TextResponseDateis assigned as DateOverall Ratingis assigned as Score
When a column header does not identify a supported data type, the column defaults to String.
Review every field carefully, even when a data type has been assigned automatically.
Your file must contain exactly one field assigned as Text. This is the open-ended feedback that Journey Insights Hub analyzes.
For metadata fields, the Data Summary panel also provides a preview of how the values will appear when used as filters after the text analysis project is created. Selecting an item in the Data Summary panel moves the Data Preview to the corresponding column and highlights it.
Review the data preview
The Data Preview panel displays up to the first 100 rows of your dataset.
Use the preview to confirm that:
Values appear under the correct column headers
The Text field contains the feedback you want to analyze
Dates, scores, and numbers fields are assigned correctly
Metadata values appear as expected
Selecting a column in the Data Preview highlights the corresponding item in the Data Summary panel.
Resolve errors and review warnings
The top of the Data Summary panel displays the number of Errors and Warnings found in the dataset.
Errors
Errors must be corrected before you can create the text analysis project. The Create project button remains unavailable while any errors are present. Text analysis project-level errors appear beneath the error and warning totals. Errors associated with a specific field appear within that field’s item in the Data Summary panel.
Warnings
Warnings identify items that should be reviewed but do not prevent you from creating the text analysis project.
For example, a warning may appear when a field’s formatting or assigned type does not match the values found in the column.
Error and warning totals update as you correct issues or change field types.
Create the text analysis project
Before completing the upload, confirm that you have:
Selected the correct workspace and language
Reviewed every field’s data type
Corrected all errors
Reviewed any remaining warnings
You can also choose whether to receive an email when the analysis is complete and the text analysis project is ready. This option is selected by default.
Select Create project to begin uploading and analyzing the data. Journey Insights Hub will process the records and build the text analysis project using the language and field settings you selected.
Adding data to an existing text analysis project
To add records to a text analysis project that has already been created, select Upload data from the Highlights page or from its tile on the Projects page.
The review process is similar to creating a new text analysis project. When the new file’s column headers do not match the existing fields, Journey Insights Hub displays warnings so you can review the differences before continuing.
For detailed instructions, see article Manage and update your text analysis project.