Extraction Guide

Learn how to get the most from your extraction prompts — from simple one-liners to detailed, repeatable workflows

Let AI Write Your Prompt

Not sure where to start? Let our AI analyze your documents and generate an extraction prompt for you.

Upload and describe your goal

Drop your files, optionally describe what you need, and click:

Click for AI to analyze files and suggest prompt

AI generates your prompt

The AI examines your documents and the goal you described to generate a tailored extraction prompt.

Review and run

Run the prompt as-is, or adjust it to get exactly what you need. Save it to your library for reuse.

Describe Your Goal for Smarter Results

Without a goal

“Extract invoice number, date, and total”

With a goal

I'm processing supplier invoices for payment. Extract invoice number, date, and total”

Why this helps

Describing your goal — such as the finance process you're performing — helps the AI make smarter decisions about edge cases in your documents (like how to handle bundled documents or ambiguous values) so you don't have to anticipate every scenario in your prompt.

More examples:

I'm processing these invoices for payment approval...I need this for our quarterly VAT return...I'm doing a line-item spend analysis across vendors...I'm reconciling these bank statement transactions against our invoices...I'm extracting monthly utility charges across our sites for cost reporting...I'm preparing payroll data for our monthly pay run...

Start simple. Add detail when you want precise control over edge cases and formatting

When your prompt leaves something unspecified, the system uses conservative, accounting-friendly judgment.

Simple Prompts

AI Handles the Details

The AI interprets your goal and documents to select relevant fields and formats.

Extract invoice number, invoice date, vendor name, net amount, tax, total

AI returns exactly these fields, correctly formatted

I’m processing invoices for payment. Extract invoice number, date, vendor, amount due, payment terms

Goal helps AI handle edge cases; listed fields define the output

Extract line items: description, quantity, unit price, line total

AI creates one row per line item with invoice details repeated

Detailed Prompts

Repeatable Workflows

Define exact fields, formats, and business rules for repeatable results and correct edge-case handling.

I'm preparing AP data for our month-end close. Extract: Invoice Number (alphanumeric, top-right) Invoice Date (YYYY-MM-DD) Vendor Legal Name (prefer extracting from footer) Net Amount (pre-tax invoice total) VAT Rate (if no VAT is listed use 0, use Excel type percentage) VAT Amount (If no VAT is present use 0) Total Amount (invoice final total) Document Type (classify as Invoice or Credit Note) - For credit notes prefix Invoice Number with 'CR-' and show amounts as negative. - One row for each invoice or credit note. - Skip any pages that are email cover sheets or summary pages”

Save & Reuse with the Prompt Library

Save prompts for one-click reuse across future batches.

One prompt or many?

You don't need separate prompts just because documents come from different vendors or have different layouts — the AI adapts to these variations. But when the extraction logic itself differs, separate prompts keep things clean: different tasks (VAT reporting vs. expense tracking), different document types (invoices vs. bank statements), client-specific output formats, or vendor-specific handling (special rules for a specific supplier).

After Extraction: AI Notes

When the AI has to make an assumption, it tells you what it assumed — and suggests how to make your prompt more explicit next time. No notes means no assumptions were needed.

Multiple possible matches

Your prompt says ‘Total’ but the document has both a line item total and an invoice total — the AI tells you which one it used.

Inexact field names

Your prompt says ‘Net’ but the document labels it ‘Subtotal’ — the AI tells you how it matched them.

Mixed document types

Your files contain invoices with attached purchase orders — the AI tells you which pages it extracted from and which it ignored.

Unspecified scenarios

Your prompt didn’t mention credit notes but the AI encountered one — it tells you how it handled it, such as treating amounts as negative.

Example AI Notes

AI Extraction Assistant Notes

Post-extraction report · 1 observation · 2 prompt suggestions

During this extraction, I made some interpretations based on your instructions and the documents provided. Review these notes to ensure the results match your expectations.

1

Documents to extract from

Your files often contain a ‘Tax Invoice’ with an attached ‘Delivery Note’. I treated the ‘Tax Invoice’ pages as the main source of data, and ignored the attached ‘Delivery Note’ pages as supporting context.

Suggested prompt additions

To confirm this handling:

Extract from ‘Tax Invoice’ only

To extract from both:

Extract from ‘Tax Invoice’ and ‘Delivery Note’

Prompt Controls & Capabilities

Define exact fields, formats, and rules to control how your data is extracted and structured.

Field Selection & Column Naming

Define exactly which fields to capture and how to name the columns.

Extract 'Invoice Number', 'Invoice Date', and 'Total'

Extract the 'Vendor Legal Name' and use the column header 'Supplier_Name'

Only these fieldsorInclude related tax fields

Output Structure & Layout

Control what each row represents, how columns are ordered, and how data is grouped.

Create one row per invoice

Create one row for each line item, and repeat the 'Invoice Number' on each row

Join all line item descriptions into a single cell, separated by a semicolon.

Order columns as: Date, Vendor, Invoice Number, Total

Business Logic & Rules

Set hints, default values, and conditional logic to handle real-world variations.

HintThe 'Product Code' is in the 'Description' column, and it always begins with 'SKU-'

DefaultIf 'Tax Amount' is missing, set its value to 0

FallbackFind the 'PO Number' in the header. If it is not present, extract it from the 'Reference' field.

ConditionalIf 'Currency' is 'USD', extract 'Tax' from the 'State Tax' field. If 'Currency' is 'EUR', extract 'Tax' from the 'VAT' field.

ConditionalIf 'Vendor' is 'Acme Corp', set 'Internal_Code' to 'ACME-001'

Document & Page Handling

Apply rules to specific document types or filter out unwanted pages.

Ignore any pages where the title is 'Email Cover Sheet'

For credit notes, prefix the 'Invoice Number' with 'CR-' and show all amounts as negative

On pages identified as a Statement of Account, extract each invoice listed in the summary table as a separate row

Data Standardization & Formatting

Control how dates, numbers, and currencies are stored in Excel — ensuring data is in the correct format for calculations, pivot tables, and system imports.

Format all dates as YYYY-MM-DD

Ensure all currency fields have 2 decimal places

Set the 'Invoice Date' column as a text type, not a native date

Note: Your local Excel settings may display native Excel types (i.e. numbers, dates) according to your settings.

Data Classification & Enrichment

Automatically categorize and enrich transactions based on contextual clues — simplifying expense tracking, departmental budgeting, and tax preparation.

Add an 'Expense Category' column. Based on the line item description, classify each item as one of the following: 'Office Supplies', 'Software & Subscriptions', 'Travel & Entertainment', or 'Utilities'.

Add a 'Payment Priority' column. If the invoice is overdue or due within 7 days, classify as 'Urgent'. Otherwise classify as 'Standard'.