Use the receipt processing prebuilt model in Power Automate (preview)

[This topic is pre-release documentation and is subject to change.]

Important

To use AI Builder models in Power Automate, you have to create the flow inside a solution. The steps below won't work if you don't follow these instructions first: Create a flow in a solution.

  1. Sign in to Power Automate and select the My flows tab.

  2. Select New, then select Instant—from blank.

  3. Name your flow, then select Manually trigger a flow under Choose how to trigger this flow.

  4. Select Create.

  5. Expand Manually trigger a flow, then select + Add an input.

  6. Select File as the input type, then set My receipt as the input title.

    Trigger file flow

  7. Select + New step, search for AI Builder, and then select Process and save information from receipts in the list of actions.

  8. Specify the My receipt field from the trigger in the Receipt file input.

  9. In the successive actions, you can use any of the receipt values from the model output.

    Flow example

Note

Receipt values are returned as strings. To manipulate them as numbers, you can use the float or int conversion functions.

Congratulations! You've created a flow that uses the AI Builder receipt processing model. Select Save on the top right, and then select Test to try out your flow.

Parameters

Input

Name Required Type Description
Receipt file Yes string The receipt file to process

Output

Name Type Description
Merchant name string Merchant name
Merchant address string Merchant address
Merchant phone number string Merchant phone number
Transaction date string Transaction date
Transaction time string Transaction time
Purchased item name string Purchased item name. Returned as a part of a list of items.
Purchased item quantity string Purchased item quantity. Returned as a part of a list of items.
Purchased item price string Purchased item price. Returned as a part of a list of items.
Purchased item total price string Purchased item total price. Returned as a part of a list of items.
Subtotal string Subtotal
Tax string Tax
Tip string Tip
Total string Total
Confidence of merchant name float How confident the model is in its detection
Confidence of merchant address float How confident the model is in its detection
Confidence of merchant phone number float How confident the model is in its detection
Confidence of transaction date float How confident the model is in its detection
Confidence of transaction time float How confident the model is in its detection
Confidence of purchased item name float How confident the model is in its detection. Returned as a part of a list of items.
Confidence of purchased item quantity float How confident the model is in its detection. Returned as a part of a list of items.
Confidence of purchased item price float How confident the model is in its detection. Returned as a part of a list of items.
Confidence of purchased item total price float How confident the model is in its detection. Returned as a part of a list of items.
Confidence of subtotal float How confident the model is in its detection
Confidence of tax float How confident the model is in its detection
Confidence of tip float How confident the model is in its detection
Confidence of total float How confident the model is in its detection
Detected text string Line of recognized text. Returned as a part of a list of text.
Page number of detected text integer Which page the line of recognized text is found on. Returned as a part of a list of text.
Height of detected text float Height of the line of text. Returned as a part of a list of text.
Left position of detected text float Left position of the line of text. Returned as a part of a list of text.
Top position of detected text float Top position of the line of text. Returned as a part of a list of text.
Width of detected text float Width of the line of text. Returned as a part of a list of text.

Receipt processing overview