Workflows · 2 min read

AI Extract Data Action

The AI Extract Data action takes unstructured text and pulls out the fields you define, such as a name, email, phone number, or order ID. The results become custom values for later steps. For example, you can turn a lead alert email into a new contact and opportunity.

Settings

| Field | What it does | | --- | --- | | Action Name | A name for the step. The default is "AI extract data." | | Extract From (required) | The text to read. Pick a custom value, such as an email body, webhook payload, SMS message, or output from an earlier step. | | Additional Context | Optional hints for the AI, such as "This is a lead notification email. Phone numbers may appear in different formats." | | Templates | Ready-made field sets you can use as they are or edit. | | Data Fields | The fields to extract. Each has a name, a type (Text, Email, Phone, Number, or Date), and an optional description. |

Templates

Contact info: full_name, email, and phone.
Opportunity info: company_name, job_title, budget, and timeline.
Order info: order_id, items, total_amount, and order_date.
Appointment info: appointment date, time, location, and type.

You can also build your own field set from scratch.

Set it up

In your workflow, click + and add AI Extract Data.
In Extract From, use the custom value picker to choose the text to read.
Add any Additional Context that helps the AI.
Pick a template or add your own data fields. Give each a clear name, type, and description.
Save the action.

Use the extracted data

In a later step, click into a field, open the custom value picker, choose AI extract data, pick this action by its name, and select the field. For example, use full_name in an Update Contact action or to personalize an email.

Tips

Use field descriptions to say exactly what each field should hold.
Test with real examples of the text before publishing.

Did this answer your question?

No, I need help