MPS Intellivector – Next generation document processing

February 26th, 2015 → 10:06 am @

MPS Intellivector – Next generation document processing

Companies these days are waging a costly battle against unorganized paper and digital documents clogging their workflows. Considerable time and money are spent on developing and integrating new solutions which often require internal resources to process these documents, lowering overall work efficiency. These solutions are anything but flexible and require further investments or even full replacement anytime something changes in the usual work process.

Now imagine a solution that can save time, money and resources by speeding up the data entry and verification process, present outsourcing options, while still guaranteeing 100% information confidentiality and high data accuracy.


  • Value add even to existing data capture solutions
  • Save up to 90% on processing cost
  • Speed up processing by up to 60%
  • 100% confidentiality of sensitive data at every step of the process
  • Support of special characters (full 3-byte)
  • Securely outsource data entry
  • Simple integration with popular enterprise systems and capture software


  1. Automated document type recognition (document-fingerprinting)
  2. Breaking down the document into small individual data fields (“Snippets”)
  3. Automated recognition of snippets (OCR, ICR)
  4. Manual data entry of the snippets
  5. Automated verification by cross-checking the results of step 3 and 4
  6. Post-processing (adding metadata) and exporting results directly into enterprise back-end systems


MPS IntelliVector breaks down the incoming images of documents or forms into small individual fields („snippets”), like part of a name or an address, or a few digits of a number. These snippets are automatically processed and, at the same time, they aresent out to users for manual data entry. These users receive snippets of information out of their original context, meaningless on their own, this way the confidentiality of the initial data is preserved. Moreover, since the Data Entry users enter the data through a specially designed, simple to use interface, and work with small strings of information only a few characters long, instead of full page images of text, the data entry process is general approximately 50% faster.


MPS IntelliVector supports special character data entry, such as Arabic, Hindi, simplified Chinese, Japanese, Thai, Tamil, Cyrillic, and so on.

With this approach companies can spare their own, expensive internal resources and outsource the whole data entry process to any part of the world realizing massive, up to 90% cost savings, without risking the confidentiality of the initial information. The results of the automated recognition and the manual data entry are crosschecked for result verification, leading to more accurate data recognition and only a fraction of the usual results (about 1%) that require additional manual Quality Control on a supervisor level.


  • Higher data accuracy by combining automated recognition and manual data entry
  • Faster data entry (data entry based on snippets vs. full page document image)
  • On-the-fly addition of new form templates (no need for IT)
  • Support of special characters
  • Simplified and fast UI – no special training required
  • Multilingual interface (English, Arabic, Russian, Thai, Malaysian, Vietnamese, etc..)
  • Work distribution (guest users) and gamification for additional motivatio