Template-based data extraction engine

Our template engine is the legacy way to extract data in Parseur. It still works, it is still supported, and it is the right tool when you want exact control over one fixed format. New mailboxes run on the AI engines instead, with no template to build.

Templates are our legacy engine

Parseur runs on AI now. The Text AI engine reads emails and text documents, the Vision AI engine reads PDFs, scans and images, and both pull the fields out on their own. There is no template to build and none to repair when a sender changes their layout. Four hundred suppliers do not become four hundred templates to maintain.

The template engine described below is still here and still supported. Keep it for the one format you want to control down to the character. For everything else, start with AI.

Wait. What does a template look like?

A parsing template is a visual representation of your document where you highlight the data that needs to be extracted. It was the original way to teach Parseur a document, and it is still easier to set up than the parsing rules other products make you write. The AI engines skip this step entirely and find the fields on their own.

Documents with different layouts? Let the AI take those.

Similar documents arriving from dozens of sources, each with its own layout, is the exact case the AI engines were built for. They read a layout they have never seen before without being taught it first. If you still want templates for a handful of fixed senders, the engine handles multiple layouts in one mailbox.

Multi-templates by default

Create as many templates as you need in your mailbox, one per layout. There is no need to create separate mailboxes for documents with different layouts and complex routing rules to send the right document to the right mailbox. The AI engines need neither.

Automatic layout detection

When there are more than one template in a mailbox, Parseur will automatically pick the right one whenever a new document comes in. There is no need to set up anything to tell Parseur how to pick the correct layout; it will do it automatically.

Zero templates anyone?

Parseur has a built-in library of managed templates for specific industries. We call it zero-template parsing. Just send us your documents, and if we have a matching template, they will be processed automatically. We build and maintain these ones, so there is nothing for you to repair when a platform changes its emails.

Supported industries include:

Real estate

Extract contact details from leads from major real estate providers around the world, including Zillow, StreetEasy, and Apartments.com.

Food ordering

Parse email or PDF orders from most food ordering platforms, including Doordash, Grubhub, Toast, and Slice.

Google Alerts

Monitor your Google Alerts closely by exporting them to Google Sheets or any other application instantly and automatically.

Job applications

Extract data from email applications from LinkedIn or Indeed.

Hotels and short-stay bookings

Parse booking confirmations from Airbnb, VRBO, and more.

Zonal and Dynamic OCR for Ultimate Data Extraction

Zonal and Dynamic OCR are template features, for scanned documents you want to map by hand. The Vision AI engine reads the same scans with no template at all, and our OCR accuracy is the same either way.

Zonal OCR

With Zonal OCR, extract text from fields that are at a fixed position on every similar document.

Dynamic OCR

With Dynamic OCR, easily extract text from fields that move horizontally, vertically or change size from one document to the next.

Best-in-class OCR software

Parseur's OCR accuracy is the best on the market. It supports most languages, including handwritten and is blazingly fast.

Get started

Ready to automate your
document data extraction?

Start free in minutes and see how Parseur fits into your workflow.

No model training required
Automates data entry from any document
Scales from point-and-click to API