- 2 Minutes to read
How to Create a Data Source
- 2 Minutes to read
In the first step of our quickstart tutorial, we will guide you through creating a data source, allowing you to choose the specific data you want to extract from your service.
Before starting, make sure your account (e.g. Facebook account or GA4 account) has sufficient permissions to access and extract your data. Usually at least admin-level permissions are required.
1. Choose the Connector
- In the Dataddo app, click on the Create a Source button under Quick Actions.
- Select your service.
2. Select Your Dataset and Account
- If your connector is dataset-based, select your dataset.
- Click on Add New Account at the bottom of the drop-down and authorize the connection of your service to Dataddo.
If the service authorization requires more information, consult our connector-specific how-to guides.
3. Select Metrics, Dimensions, or Attributes
Select your metrics, dimensions, or attributes to choose what data to extract. In short, you are defining the actual data model of the source (i.e. what columns will be included in the source).
4. Configure Snapshotting
- Select the frequency of your snapshotting, aka how often your data should be extracted. Recommended default is daily snapshotting.
- In the advanced settings, you can choose your date range and at what time data should be extracted.
For loading historical data, refer to the Data Backfilling page.
5. Test the Configuration and Save Your Data Source
- Click on the Test Data button to preview and confirm your data.
- Save your configuration. Congratulations, your new data source is ready!
Test Data is also a great feature for data exploration when you are unsure about what certain attributes mean. As your table will be populated with live data, it will be easy to discover values for each attribute.
Dataset-Based vs Non Dataset-Based Connectors
Dataset-based connectors offer pre-selected set of attributes, while non dataset-based connectors allow complete freedom to compose your data source.
If you are not sure which dataset you need but know the metrics and attributes, use the Search by Name or Attribute function.