Two Ways to Parse the Same Invoice, and Only One Survives a Layout Change

Описание и применение

Использование AI Document Parser позволяет извлекать данные из счетов-фактур независимо от изменений макета, в отличие от традиционных шаблонов с фиксированными областями захвата, которые ломаются при изменении расположения полей.

Задача

Finance teams rely on PDF parsing templates that break when vendors change invoice layouts, causing data extraction failures.

Результат

Using the AI Document Parser eliminates the need for fixed capture areas, making invoice extraction resilient to layout changes and reducing maintenance effort.

Рабочий процесс

• 1. Create a Parse Document template: upload a sample PDF, draw capture areas, add regex rules, save to get a TemplateId.
• 2. Call the Parse Document REST endpoint with the TemplateId to extract data.
• 3. Observe failure when the vendor changes layout (capture areas no longer match).
• 4. Create an AI Document Parser Analyzer: define a JSON schema for required fields.
• 5. Call the AI Document Parser endpoint (or use no‑code connectors) with the Analyzer Id.
• 6. Extract data reliably even after layout changes.
• 7. Optionally use pre‑tuned AI Invoice Parser for standard invoices or Classify Document to route mixed inboxes.

Доказательства

The AI reads the document semantically against that description instead of checking a fixed position on the page, which is exactly why a shifted total doesn’t break it the way it breaks a capture‑area template.

Источник

Zapier's own field for this is labeled "Template Name," but the value it actually wants is the TemplateId GUID, not the human‑readable name typed in when the template was created.

Источник