Set Up Intelligent Document Reader

Learning Objectives

After completing this unit, you’ll be able to:

Explain the challenges of paperwork and how Intelligent Document Reader simplifies document handling.
Create credentials.
Enable Intelligent Document Reader.
Enable image previews for documents.
Create new document types and templates.

Challenges with Paperwork

Paperwork, although resource-draining, is essential in administrative jobs across sectors. Manual data entry is inseparable from daily operations in healthcare, banking, education, and many other industries. Administrative workers must review documents; extract, verify, and enter data; and process information with high accuracy. A significant downside of manual data handling is that it’s prone to human errors. With businesses getting busier and the competition stiffer, resource optimization and data automation are top priorities today.

Bloomington Caregivers is a home healthcare agency for the elderly in Bloomington, Minnesota. With the agency’s growing popularity and increasing workload, the team is struggling to process the huge inflow of documents. A major concern is the lack of scalable solutions to match the administrative load on care coordinators, doctors, and nurses, which is undermining the care quality.

Enter Intelligent Document Reader

Intelligent Document Reader is a Salesforce solution that eliminates manual data entry and minimizes errors in data handling. Using optical character recognition with Amazon Textract, it extracts data automatically from documents and enters the extracted data into Salesforce records. This saves the time that businesses otherwise invest in managing high document inflow and processing data manually.

Intelligent Document Reader works with PDF, JPG, and PNG image files; supports six languages (English, Spanish, German, Italian, Portuguese, and French); and can even read documents handwritten in English.

Users specify the kind of document they want Intelligent Document Reader to extract data from, and then map fields in the document to fields in Salesforce objects. The extracted data can be used to create or update Salesforce records, or to verify data that’s already there in the org.

Harryette Randall, the Salesforce admin at Bloomington Caregivers, turns to Intelligent Document Reader to ease the burden of the team. For a start, she decides to use it for handling patient referrals. In this module, find out how the Bloomington Caregivers team sets up and uses Intelligent Document Reader.

Licenses and Setup Requirements

Intelligent Document Reader is available with Automotive Cloud, Consumer Goods Cloud, Education Cloud, Financial Services Cloud, Health Cloud, Manufacturing Cloud, Media Cloud, Net Zero Cloud, and Public Sector Cloud. To use Intelligent Document Reader, users need the Intelligent Document Reader add-on license.

In this module, we assume you’re a Salesforce administrator with the required permissions to set up Intelligent Document Reader. If you aren’t one, that’s OK. Read along to learn how your admin would take the steps in a production org. Don’t try to follow these steps in your Trailhead Playground. Intelligent Document Reader isn't available in the Trailhead Playground.

As Intelligent Document Reader uses Amazon Textract to extract data from documents, Harryette creates an Amazon Web Services (AWS) account and assigns the required permissions.

To set up Intelligent Document Reader, Harryette must do the following tasks.

Create an external credential, a named credential, and a permission set to authenticate the AWS account.
Enable Intelligent Document Reader and assign object permissions.
Enable image preview for documents.
Create a document type.
Create a template and field mapping.

Let’s follow along.

Create Credentials

Here’s how Harryette creates an external credential.

Click and select Setup.
In the Quick Find box, search for and select Named Credentials.
Click External Credentials.
Click New.
Enter Label: Demo Credentials
Enter Name: Demo_Credentials
Select Authentication Protocol: AWS Signature Version 4
Enter Service: textract
Enter Region: us-west-2
Click Save.
On the Demo Credentials page, scroll down to Principals and click New.
Enter a parameter name and ensure that the Sequence Number field has a value.
Click Save.

Harryette then creates a named credential for the external credential. And she creates a permission set named TextractAccess to use the named credential.

Enable Intelligent Document Reader

Next, Harryette enables Intelligent Document Reader in the org.

Click and select Setup.
In the Quick Find box, enter Intelligent Document Reader and select Intelligent Document Reader.
Click Use Your Own AWS Account.
Click Accept & Continue.
Select the named credential that you created and click OK.
In Global Content Extraction Settings, click Edit Settings.
In the Maximum Pages field, enter 5.
In the Confidence Score Threshold field, enter 80.

The Intelligent Document Reader enabling window.

Here’s what the options Maximum Pages and Confidence Score Threshold mean.

Maximum Pages: The maximum number of consecutive pages from the beginning of a document that can be sent for content extraction. The default value is 5. If pages that are already scanned are selected, the pages aren’t counted against the limit.

Confidence Score Threshold: Indicates the reliability of the data in a document. Intelligent Document Reader returns a confidence score ranging from 0 to 100 for the information it extracts from any document. Record fields scoring below the threshold are highlighted.

To make sure users can use Intelligent Document Reader, Harryette gives them object permissions.

Enable Image Preview for Documents

Harryette enables the image preview option so Intelligent Document Reader can preview images of documents.

Click and select Setup.
In the Quick Find box, enter Salesforce Files and select General Settings.
Click Edit.
Select Display alternative file previews.
Click Save.

The General Settings window within Salesforce Files Settings.

Create a Document Type

Harryette must create a document type for each document that she wants to extract information from. She creates a document type that she’ll use for handling patient referrals.

Click and select Setup.
Enter Document Type in the Quick Find box and select Document Type.
Click New Document Type.
In the Label field, enter Demo Docs. The Name is automatically set to Demo_Docs.
Select Other Forms for Form Type. If your document is a US license or passport, select Identity Document.
Enable Is Active if it’s not selected.
Click Save.

The New Document Type window.

Create a Template and Field Mapping

Finally, Harryette creates a template and a mapping to accommodate documents related to patient referrals.

In the Quick Find box, enter Intelligent Document Reader and select Intelligent Document Reader.
Click New Template.
In the Name field, enter Patient Referrals. The API Name is automatically set to Patient_Referrals.
For Document Type, select Demo Docs.
Click Upload File and select the standard patient referral document. You can upload a maximum of five files when creating templates for document types.
Click Add (1).
Click Continue.
For Target Objects, select Lead (Unmapped). Optionally, you can also map form fields to the selected object’s record type.
Click the uploaded patient referral document and map Document Fields to Target Object Fields.
Once all the document fields are mapped to their target objects, click Save.
For any field in the document that Intelligent Document Reader fails to extract, click Add Field to add it manually.
Click Activate.

Harryette has successfully set up Intelligent Document Reader to handle patient referrals at Bloomington Caregivers. She has created custom fields to capture relevant information from uploaded documents and accommodate the data in new documents. In the next unit, find out how the team uses Intelligent Document Reader to manage data handling tasks.