Blog > Seeing is believing: The power of MetaServer in action

Reading about an Intelligent Document Processing (IDP) solution is one thing. Watching it untangle a stack of mixed documents, detect and extract messy handwriting or compress a color PDF tenfold in real time is another. Seeing is believing and that's why we've put together a library of short, practical demo videos showing exactly what MetaServer can do with real-world documents.

If you're evaluating IDP software for your organization, these videos give you something most product pages can't: proof. Below, we walk through all eight demos so you know exactly what to expect.

We recorded MetaServer working on the same kinds of documents our customers deal with every day, delivery notes scanned on a cheap mobile scanner, attorney letters with inconsistent layouts, attendance lists with eleven different handwriting styles and much more.
Each video runs just a few minutes and focuses on a specific capability.

This is just the tip of the iceberg, MetaServer has so much more capabilities than just the 8 demo videos discussed here.
We just highlighted a couple of them to show the power of our IDP solution.

You can watch all eight videos on our MetaServer demos page.

The first demo shows MetaServer automatically classifying documents using Azure AI Document Intelligence. After training a custom classification model with just five sample documents (a simple drag, drop and tag process), MetaServer can take a single PDF containing several mixed document types and instantly separate and identify every document inside it. No separator sheets and no manual sorting required.

Low-confidence results are flagged in orange for validation and those flagged files automatically feed back into the model's training set, so classification accuracy keeps improving over time.

Real-world documents are rarely pristine. This demo compares a delivery note scanned on a high-end office scanner against one captured on a basic mobile scanner in the cabinet of a truck. MetaServer detects an extracts accurate data from both.

This is the difference between traditional OCR and Intelligent Document Processing. MetaServer, powered by AI document technology, interprets content rather than simply recognizing characters. 

Our rule of thumb is simple: if a human can read it, MetaServer can read it.

Some of the hardest documents to automate are the ones where the same data points (file numbers, dates, names) appear in different locations depending on who created the document. This demo uses attorney mortgage redemption letters as an example, a notoriously difficult document type because every law firm formats them differently.

By training a custom extraction model through simple click-and-tag sampling, MetaServer learns to find that "floating" data regardless of layout.
This including printed text, handwriting, barcodes and tables that span multiple pages.

The result: unstructured documents go in, structured data comes out.

Handwritten documents have traditionally meant manual data entry due to the limitations of OCR.
This demo shows MetaServer reading an attendance list with eleven handwritten names, each written by a different person in a distinct style.

Powered by Azure AI Document Intelligence, MetaServer handles cursive, mixed formats and multiple languages easily.
Confidence scoring flags any uncertain entries for a quick human check, so manual transcription becomes unnecessary.

Validation is what makes automation trustworthy. This demo walks through MetaServer's validation client, where operators verify, correct or complete extracted data before it moves further down the workflow.

Unlike many LLMs and AI tools that lose visual context once data is extracted, MetaServer keeps a side-by-side view of the extracted value and its exact location on the original document, so validating data is fast and reliable.

A standard image-based PDF is only searchable if the text was already digital. This demo shows MetaServer converting scanned documents into fully searchable PDFs. The result is a document where every word, printed or handwritten, can be found with a simple search.

Bandwidth and storage efficiency matter just as much as accuracy, especially for cloud-based workflows. This demo introduces MetaServer's Convert to PDF MRC (Mixed Raster Compression) module, which compresses color PDFs by a factor of 10 while keeping text, stamps and highlighted details crisp and legible.
A side-by-side comparison in the video shows just how little quality is lost.

The final demo covers MetaServer's Digital Imprinter module, which embeds metadata such as processing dates, operator names or dynamic QR codes directly onto documents. Imprints can be applied to specific pages or a custom header sheet and a built-in preview lets you verify the layout before rolling it into production. This is a simple way to build traceability directly into the documents themselves.

Every one of these videos focuses on a real capability our customers rely on daily: classification, extraction, validation, searchable PDF conversion, compression, and traceability. Together, they show how MetaServer moves beyond basic OCR into genuine Intelligent Document Processing.

Ready to see it for yourself? Watch all eight demos on the MetaServer Demos page or get in touch if you'd like a live walkthrough tailored to your own documents.