Classify and Parse Documents using PDF.co and Make

Document processing is a fundamental task for many organizations, and managing large volumes of data can be a hard task. Two important processes that can help organizations simplify their document processing workflows and extract valuable insights from their data are classifying and parsing.

PDF.co and Make are two powerful tools that offer a range of features for automating document processing tasks, including the ability to extract data from PDFs, images, forms, tables, and more. By using these tools, organizations can convert unstructured data into a structured format that can be easily analyzed and processed, leading to significant cost savings and increased productivity.

In this guide, we will explore the process of classifying and parsing documents and how they can help organizations extract valuable insights from their data with the help of PDF.co and Make.

1. Create a New Scenario
2. Add Google Drive Module
3. Setup Google Drive Configuration
4. Google Drive Result
5. Add Iterator Tools
6. Setup Flow Control
7. Add PDF.co Module
8. Setup PDF.co Configuration
9. Document Classifier Result
10. Add Set Variable Tools
11. Setup Tools Configuration
12. Add Another PDF.co Module
13. Configure PDF.co Settings
14. Parse Document Result
15. Add Google Sheets Module
16. Setup Google Sheets Configuration
17. Run Google Sheets Result
18. Google Sheets Output
19. Demo

In this document processing example, we will use a set of sample PDF documents to demonstrate how automation tools like PDF.co and Make can be used to extract data and classify the extracted information from an invoice.

Step 1: Create a New Scenario

• Let’s begin by logging into your Make account and creating a new scenario.

Step 2: Add Google Drive Module

• Next, search and select the Google Drive app. Then, choose the Watch Files in a Folder module to trigger when a file is created or modified in a selected folder.

Step 3: Setup Google Drive Configuration

Let’s set up the Google Drive configuration.

• Begin by connecting your Google Drive account to Make. This will allow Make to grant access to your Google Drive account and enable you to fetch files from your Drive.
• In the Watch Files section, select By Created Time as the trigger for the automation. This means that Make will fetch files from the specified folder based on the time they were created.
• Choose the Drive where the folder you want to fetch is stored. This is the Google Drive account where the folder containing the files you want to fetch is located.
• Select the folder you want to watch in your Google Drive account. This is the folder where Make will fetch the files from.
• Specify the file types you want to watch for. This is the type of files you want Make to fetch from the specified folder in your Google Drive account.
• Enter the maximum number of results you want to work with during each execution cycle. This allows you to control the amount of data being processed, ensuring that the automation runs smoothly without overloading the system.

Once you have set up the configuration for fetching files from your Google Drive in Make, you can run the scenario to automate the process.

Step 4: Google Drive Result

• Great! The scenario runs successfully and detects new files to trigger the automation to parse the invoices. Let’s add another app to classify the extracted information from an invoice.

Step 5: Add Iterator Tools

• In this step, we will enhance the functionality by incorporating Iterator tools, which enable us to execute a series of commands iteratively on a given list of items.

Step 6: Setup Flow Control

• Under the Flow Control in the Array section, select the Web Content Link from Google Drive.

Step 7: Add PDF.co Module

• In this step, let’s add the PDF.co app to the workflow and choose the Document Classifier module to analyze the text of the input document and return its class.

Step 8: Setup PDF.co Configuration

Let’s configure the PDF.co settings.

• Connect your PDF.co account to Make by adding the API Key. You can obtain the API Key from your PDF.co dashboard or by signing up at this link.
• In the Input File field, select the Import PDF or Image from URL option to utilize the iterator return value.
• In the Set Custom Rules field, define specific rules to extract data fields such as the invoice number, account number, and total amount due. By setting custom rules, you can automate the document processing workflow and ensure accurate and efficient processing of documents.

Once you have completed the PDF.co configuration setup, you can proceed to run the scenario for document classification and retrieve the corresponding class value.

Step 9: Document Classifier Result

• Excellent! We successfully classify the documents and return their class values. Now, let’s add another module to specify the name of the variable and the value to be assigned to it.

Step 10: Add Set Variable Tools

• In this step, we will add another module and select the Set Variable tools to allow us to define a value once and use it multiple times throughout your build process, which can make your files more concise and easier to maintain.

Step 11: Setup Tools Configuration

Let’s set up the configuration.

• In the Variable name field, add the name of your variable. This is the name that will be used to refer to the variable throughout the automation workflow.
• Next, specify the variable’s lifetime. This refers to the period of time during which the variable will exist in memory.
• Then, assign a value to the variable using an iterator or other appropriate method. You may use an iterator to assign a unique value to the variable for each item in a list, or you may set the variable based on user input or other conditions specified in the automation workflow.

After you have set up the variables in your Make scenario, you can execute the scenario to observe the outcome. This will allow you to test the variable setup and ensure that the variables are being assigned the correct values based on the automation workflow. Once the scenario runs successfully, add another module to parse data from the PDF invoices

Step 12: Add Another PDF.co Module

• In this step, we will add another module and select the PDF.co app. Then, choose the Parse a Document module to parse data from PDF invoices.

Step 13: Configure PDF.co Settings

Let’s set up the PDF.co configuration.

• In the Input File field, select the Import a file from URL option to utilize the Iterator return value.
• In the URL field, enter the Iterator return value.
• In the Document Parser Template ID field, enter the variable name that holds the parsed data. You can create a template ID by visiting this link.

Once you’ve finished setting up the PDF.co configuration, run the scenario to extract the parsed data value from the invoices.

Step 14: Parse Document Result

• Excellent! PDF.co has processed our request and returned the parsed data value from the invoices. Let’s proceed to save the parsed data value to Google Sheets.

Step 15: Add Google Sheets Module

• In this step, let’s integrate the Google Sheets app into the workflow and select the Add Row option to add a new row to the Google Sheets spreadsheet and map the columns in the row with the corresponding parsed data values.

Step 16:Setup Google Sheets Configuration

Let’s set up the Google Sheets configuration.

• Connect your Google Sheets account to Make by providing your Google credentials. This will allow your Make scenario to access and modify your Google Sheets files.
• Choose a path and method for accessing the spreadsheet. This can be done using the Google Sheets tool in Make, which provides options for specifying the location and method of access for the spreadsheet.
• Select My Drive as the drive to use, which will allow you to access and modify spreadsheets stored in your Google Drive account.
• Enter the Spreadsheet ID for the specific spreadsheet where you want to save the parsed data values. This ID can be found in the URL of the spreadsheet and is used to uniquely identify the spreadsheet in Google Sheets.
• Specify the name of the worksheet where you want to save the parsed data values. This can be done using the Worksheet Name field in the Google Sheets tool in Make.
• Indicate whether the table contains headers by selecting Yes or No. If the table contains headers, Make will use these headers to map the parsed data values to the appropriate columns in the worksheet.
• Finally, under the Values field, map the parsed data values from PDF.co to the appropriate columns in the worksheet.

Once you have set up the Google Sheets configuration, run the scenario to save the parsed data values to the Google Sheets spreadsheet.

Step 17: Run Google Sheets Result

• Congratulations on successfully running the scenario and saving the parsed data values to the Google Sheets spreadsheet! To view the parsed data values, open your Google Sheets app and navigate to the worksheet where you specified the data to be saved.

Step 18: Google Sheets Output

• Here are the parsed data values from invoices that were successfully saved to the Google Sheets spreadsheet.

Step 19: Demo

• Check out this PDF.co Document Classifier and Document Parser in action workflow.

This tutorial taught you how to Classify and Parse Documents using PDF.co and Make. You learned how to use PDF.co Document Classifier to classify or categorize documents based on their content. You also learned how to use the PDF.co Document Parser to automatically extract data fields from PDF documents and images.