Turn Incoming Emails into Data with the AI Email Parser
The AI Email Parser action is built for one job: reading the email that started your workflow and extracting the details you care about, such as name, phone number, budget or order number. It is a premium action with a usage charge per run, and it requires an Inbound Email trigger.
Configure the trigger first
- Create a workflow and add the Inbound Email trigger.
- Add filters so only the right emails enter: sent to a particular mailbox, sent from a specific address, subject, body text (for example, contains "New Buyer Lead"), or has attachments.
- Turn on Trigger Only for New Email Conversations so replies inside an existing thread do not run the workflow again.
- Save the trigger.
Add and configure the parser
- Click + under the trigger and choose Email Parser from AI Actions.
- Name the action.
- In Email Content, pick what to read: Body (latest message), Full Body (including the thread), or a Custom Value. Body choices automatically include the subject.
- Start from a template (New Lead, Order Details, Inquiry or Appointment) or define your own fields.
- Give each field a name and a type: Text, Email, Phone Number, Date, Date and Time, URL or Boolean (yes/no).
- Save the action.
Use the results
Extracted values are available to every following step. Add Create Contact, Update Contact Field, Create Opportunity, Assign to User, an SMS or an internal notification, and insert the parsed values with the variable picker.
Parser or Extract Data?
Choose Email Parser when the source is always an inbound email. Choose AI Extract Data when the text comes from other places, such as a webhook, an SMS or an earlier AI step.
Tips
- Use a Boolean field for yes/no details like "wants a callback".
- Test with an email from each of your sources, since layouts differ.
- Always create or update a record afterward; parsed values vanish unless you store or use them.
Good to know: Filtering tightly on sender and phrase keeps unrelated mail from consuming paid AI runs.
Was this article helpful?
