Skip to main content

​
Enrich a Contact List with the CLI

In this tutorial you will upload a contact file, normalize it into a Landbase dataset, run contact enrichment, and download the enriched child dataset. What you will need: Landbase CLI installed and authenticated, plus a CSV or Excel file containing contacts. Rows should include a LinkedIn URL, or a name and company domain.

​
1. Upload the contact file

Copy the dataset ID from the JSON response. Call it <uploaded-dataset-id> below.

​
2. Onboard the dataset

Onboarding normalizes the column names and values so later workflows can use them reliably:
Copy the output_dataset_id from the completed workflow. Call it <onboarded-dataset-id> below.

​
3. Start contact enrichment

Run the workflow on the onboarded dataset:
Email and phone enrichment run by default. A positive --enrich-phone flag selects phone-only opt-in behavior. Use --enrich-email and --enrich-phone together when you want both channels explicitly. The command returns a workflow run ID. Call it <workflow-run-id> below.

​
4. Wait for the enriched dataset

A successful run returns an output_dataset_id for the enriched child dataset. Missing email or phone values are normal when no matching data is available.

​
5. Publish and download the results

Publish the enriched dataset to a file format:
Then download the published child dataset:
Use datasets lineage <onboarded-dataset-id> --direction=children if you need to find the child dataset IDs created by the workflows.

​
Processing large datasets

Use --row-slice=START:END to process a half-open row range when you need to split a large input into separate runs. Run slices one at a time, then recombine the output datasets with workflow union without dedupe keys. Use --overwrite-existing only when the workflow should replace existing contact values.

​
What you learned

  • Contact enrichment is a dataset workflow, not a raw JSONL submission command.
  • Upload and onboard the input before running workflow contact-enrich.
  • Contact enrichment creates a child dataset and is asynchronous.
  • Use the workflow run ID to recover status and the output dataset ID to continue processing.
Next step: See How enrichment works for the shared enrichment model and contact-enrich migration reference for the removed command’s migration path.