Invoice, Delivery Note, Bank Statement: How to Pick the Right Preset for Every Document Type

If you've ever tried to extract data from a financial document, you already know that not all documents are created equal. An invoice has a completely different structure than a bank statement, and a delivery note requires different fields than a receipt photo. Choosing the wrong extraction approach means wasted time, manual cleanup, and errors that sneak into your spreadsheets.

Tablola solves this with presets — ready-made AI workflows built for specific document types. But with several presets available, how do you know which one to use? This guide walks you through four of the most common document types and maps each one to the preset that handles it best.

1. Invoices: Structured but Surprisingly Varied

Invoices seem straightforward — a seller, a buyer, some line items, a total. But in practice, every supplier formats their invoices differently. Some list taxes inline, others add them at the bottom. Some include product codes; others don't. A generic PDF-to-Excel tool will often miss subtleties like multi-line item descriptions or split tax columns.

The Invoice to Excel preset is purpose-built to recognise invoice structure regardless of formatting variations. It reliably extracts vendor details, invoice number, date, line items with quantities and unit prices, tax breakdowns, and totals — all mapped into a clean, consistent spreadsheet. If you're working with Turkish invoices specifically, the Fatura Verisini Excel'e Aktarma preset applies localised field logic for VAT and e-invoice formats.

• Best for: Supplier invoices, purchase records, accounts payable workflows
• Output: Line-level detail with totals and tax columns
• Works with: Digital PDFs, scanned invoices, image files

2. Bank Statements: Volume, Consistency, and Date Logic

Bank statements present a different challenge. Individual transactions are simple, but there can be hundreds of them across dozens of pages — often with running balances, varying debit/credit conventions, and merged cells in scanned versions. You need every row, in the right order, with the right sign on every amount.

The Bank Statement to Excel or CSV preset is optimised exactly for this: it handles high-volume tabular data, correctly interprets debit and credit columns, preserves transaction dates, and outputs a clean ledger-style spreadsheet ready for reconciliation or further analysis. Whether you're working with a PDF export from your banking portal or a scanned paper statement, this preset maintains row integrity across the full document.

• Best for: Account reconciliation, bookkeeping imports, cash flow tracking
• Output: Date, description, debit, credit, balance — one row per transaction
• Works with: Multi-page PDFs, scanned statements, CSV-adjacent formats

3. Delivery Notes and Shipping Documents: Item-Level Accuracy Matters

Delivery notes (also called dispatch notes or packing lists) are operationally critical — they confirm what was actually shipped, not just what was ordered. The fields that matter here are different: item codes, quantities, units of measure, and delivery references take priority over pricing. A mismatch between what was delivered and what was invoiced is a real business problem, and your data extraction needs to reflect that precision.

The Delivery Note to Excel preset focuses on logistics-oriented fields, extracting SKUs, quantities, batch numbers where present, and shipment references into a structured table. This makes it easy to cross-reference against purchase orders or invoices in the same spreadsheet environment.

• Best for: Goods receipt verification, warehouse records, supplier reconciliation
• Output: Item-level rows with quantities, units, and reference numbers
• Works with: Printed delivery notes, scanned documents, image photos

4. Receipt Photos: Unstructured, but Extractable

Expense receipts are the messiest document type in this group. They're often photographed at an angle, printed on thermal paper that fades, and vary wildly in layout from one retailer to the next. Manual data entry from receipts is one of the biggest time sinks in expense management.

The Receipt Photos to Excel preset uses AI to interpret unstructured receipt images — pulling out merchant name, date, item descriptions, amounts, and totals even when the formatting is inconsistent. It's designed for real-world receipt quality, not perfect scans. For teams managing large volumes of expense claims, this preset alone can save significant hours every month.

• Best for: Expense reporting, petty cash records, travel reimbursements
• Output: Merchant, date, line items, totals — normalized across varied formats
• Works with: Phone photos, scanned receipts, low-quality images

When You Have Multiple Document Types at Once

Sometimes you're not dealing with a single document — you have a folder of invoices from different suppliers, or a month's worth of receipts alongside delivery confirmations. In that case, the Merge Multiple Documents into One Table preset lets you process a batch of similar documents and consolidate the results into a single structured spreadsheet, rather than running each file individually.

Pro tip: Use the right single-document preset first to confirm your output looks correct, then switch to the batch merge workflow for production-scale processing.

Choosing Your Preset: A Quick Decision Guide

1. Invoice from a supplier? → Invoice to Excel preset
2. Bank or credit card statement? → Bank Statement to Excel or CSV preset
3. Delivery note or packing list? → Delivery Note to Excel preset
4. Expense receipt (photo or scan)? → Receipt Photos to Excel preset
5. Multiple documents to merge? → Merge Multiple Documents into One Table

The right preset isn't just about convenience — it's about getting the correct fields extracted reliably, without spending time reformatting output or hunting for missing data. Tablola's presets encode document-specific logic so the AI knows what to look for before it even starts. Pick the one that matches your document type, and your spreadsheet will be ready before you'd have finished typing the first row manually.