The Workflow Setup process walks you through creating a new Workflow, naming it, and connecting the Datasets that your Pipelines will draw from. This article covers each step from creation through to your first connected Dataset.
Creating a New Workflow
There are two ways to create a new Workflow:
From the home screen, click New Workflow in the Quick Actions panel.
From the top navigation, select Build → Workflows, then click Create Workflow in the upper right.
Clicking New Workflow from the home screen Quick Actions panel
Adding a Name and Description
After clicking New Workflow or Create Workflow, a modal will appear prompting you to enter your Workflow details before it is created.
Name (required): Enter a clear, descriptive title. This name identifies the Workflow across the platform and in your team's activity history.
Description (optional): Use this field to provide context about the Workflow's purpose, data sources, or use case. A good description helps teammates understand what the Workflow does in environments with multiple Workflows running.
Click Create to confirm and open the Workflow.
The Create New Workflow modal with Name and Description fields
👉 You can rename a Workflow at any time after creation by clicking the edit icon (pencil) immediately to the right of the Workflow name at the top of the page.
Connecting Datasets
Once your Workflow is created, you will be taken to the Data tab. This is where you connect the Datasets that will serve as data sources for your Pipelines.
The Workflow Data tab showing the prompt to connect a first Dataset
When no Datasets have been connected yet, the Data tab displays a "Get started by connecting a Dataset" prompt with a large Add Dataset button in the center of the page. Click it to open the Dataset search panel.
Once one or more Datasets are connected, an Add button appears in the upper-right corner of the Datasets section to add additional sources.
What Are Datasets?
A Dataset is a defined collection of data that enters your Workflow. Datasets can include:
Structured data from applications such as Salesforce, HubSpot, or other platforms via Nexadata's Connection Manager
Semi-structured data such as spreadsheets (CSV, Excel, Google Sheets)
Unstructured data sourced from unstructured file types such as PDFs, processed via the Dataset Builder
Each Dataset is added individually and can be configured based on its source, format, and structure.
Searching for a Dataset
The Dataset search panel lets you find and select the Dataset you want to connect to.
Use the Search by name field to search for a specific Dataset.
Browse the Recently Updated list to quickly find Datasets you have worked with recently.
Each Dataset in the list displays its name, the connection source it belongs to, and a data type tag on the right:
Structured: The Dataset is organized into a defined schema with rows and columns, such as a spreadsheet or database table.
Unstructured: The Dataset contains data from unstructured file types, such as PDFs.
The Dataset search panel showing recently updated Datasets with Structured type tags
Previewing and Adding a Dataset
After selecting a Dataset from the list, a Dataset Preview panel will open showing the Dataset's columns and structure. Review the columns to confirm this is the correct Dataset before connecting it.
When ready, click Add to Workflow to connect the Dataset.
The Dataset Preview panel showing columns before adding to the Workflow
👍 You can connect multiple Datasets to a single Workflow by repeating this process.
👉 Datasets must be created before they can be connected to a Workflow. If you have not yet set up a Dataset, navigate to Setup → Datasets to create one first.
Managing Connected Datasets
Once Datasets are connected, the Data tab lists all Datasets associated with the Workflow. For each Dataset you will find:
Name: The label assigned to the Dataset
Dataset Type: Indicates whether the data is Structured, Semi-Structured, or Unstructured
Actions: Options to edit, update, or remove the Dataset
👍 You can search across connected Datasets using the search bar at the top of the table.
Best Practices
Use clear, descriptive names for Datasets so they are easy to reference in Pipelines.
Ensure your source connections are tested and authorized before building Datasets.
Combine multiple Datasets to bring together related data (for example, Customers and Orders) in a single Workflow.
Next Steps
Once your Datasets are connected, the platform will prompt you to move on to building your Pipelines. Click Go to Pipelines in the next step prompt at the bottom of the page, or select the Pipelines tab from the top navigation within the Workflow.
Ready to build your Pipelines? Continue to Workflow Pipelines.