AI data extraction and email parsing

AI data extraction turns unstructured text into clean fields your workflows can use. You tell the AI what to pull out, such as a name, phone number or order ID, and it returns those values as workflow variables. That way you don't have to sort through lead emails by hand.

In HLB, AI Extract Data is a premium workflow action that works on any text you select, including SMS content, email bodies, webhook payloads, form response text and output from an earlier AI step. AI Email Parser is built for inbound emails, such as lead notifications, inquiries, appointment emails and orders, and requires an Inbound Email trigger.

To set one up:

• Add the action and choose the source. For the email parser, pick the latest email message, the full thread or a custom value.
• Add optional context, such as noting that phone numbers may appear in different formats.
• Start from a template or define your own fields. Extract Data templates cover contact, opportunity, order and appointment info. Parser templates cover new lead, order details, inquiry and appointment.
• Give each field a name, a type and an optional description. Extract Data supports text, email, phone, number and date. The parser supports text, email, phone number, date, date and time, URL and boolean.
• Add downstream actions such as Create Contact, Update Contact Field, If/Else or notifications, and map the extracted values from the custom value picker.

Extracted values exist only inside the workflow, so a later action must save them. For the email parser, you can filter the Inbound Email trigger by mailbox, sender, subject, body text or attachments, and run it only for new email conversations so replies don't trigger it again.