Bank Statement Parser Setup Guide

Before You Start

Before setting up your parser, ensure you have:

1. A PerfectParser account (sign up at perfectparser.com).
2. A healthy credit balance (each bank statement page processed consumes 1 credit).
3. A sample bank statement (PDF, PNG, or JPEG) to use during the parser setup.

Recommended Parser Settings

When creating a new Bank Statement Parser, choose the document organization structure that matches your files:

• Single Document Per File: Select this option if each statement file represents a single complete statement (even if it spans multiple pages).
• One Document Per Page: This option is rarely used for bank statements unless you scan separate single-page statement letters together in a single file.

Recommended Fields

We suggest configuring the following schema fields for your Bank Statement Parser to capture all essential ledger and summary data.

AI Instructions (Optional)

To help the AI parser extract field values with maximum accuracy, copy and paste the following prompt instructions into the AI Instructions (Optional) field during setup:

Extract the bank name, account number, statement period, starting balance, ending balance, total deposits, and total withdrawals. Also extract the transactions table containing the transaction date, description, deposit/credit amount, withdrawal/debit amount, and running balance.

Step-by-Step UI Setup

Follow these steps to set up your Bank Statement Parser using the single-page setup in the PerfectParser dashboard.

Step 1: Create the Parser

Log in to your dashboard, click Parsers in the left-hand navigation menu, and click the New Parser button.

Step 2: Configure Name and File Organization

In the Parser Details & Setup section:

1. Enter a clear, recognizable name in the Parser Name field (e.g., "Bank Statements").
2. Choose your document structure under How are your files organized? (e.g., select Single Document Per File).

Step 3: Upload your Sample Bank Statement

Upload your sample document by dragging and dropping it into the upload box on the left-hand Sample Document Preview panel, or clicking the box to choose a file from your computer.

Step 4: Run Auto-Detection

Once the document finishes uploading:

1. Copy the prompt from the AI Instructions (Optional) section above and paste it into the AI Instructions (Optional) box.
2. Click the Auto-Detect Fields button. The AI will analyze the sample document's layout and suggest extraction fields within 10 to 15 seconds.

Step 5: Verify and Refine Fields

Review the generated schema on the right. You can rename fields, change data types, or edit the helper descriptions to guide extraction. When you are satisfied with the schema, click Create Parser at the bottom right.

Step 6: Run a Batch Extraction

To process your bank statements:

1. Click Bulk Extraction in the left sidebar or the top header.
2. Select your newly created Bank Statements parser.
3. Upload one or more statement PDFs. PerfectParser will start extracting data immediately.

Step 7: Export Extracted Data

Navigate to Batch History in the left sidebar or click the Batch History tab on the parser details page. Click Show Files to inspect individual statement extractions, view confidence scores, and verify the structured output.

To download your structured data:

1. Click the Export button on the completed batch card.
2. Select Export as Excel or Export as CSV from the dropdown menu to download the structured table directly to your computer.

Developer Integration

If you want to automate bank statement parsing programmatically inside your applications:

• See the Create Parser API Guide to manage your parsers programmatically.
• See the Start Extraction API Guide to submit bank statement files and retrieve extracted JSON payloads.
• See the Webhooks Reference to configure automatic webhooks that deliver extraction results to your server as soon as they are completed.