Iron Triangle of Data Extraction

1 0

Modern businesses are fueled by data. Data drives the decision making in every facet of the business cycle. But, data is not always accessible or easy to attain.

In the mortgage industry, data is often ‘locked’ away in scanned paper documents or PDF documents.  The data exists, but not in a form that is useful. Think of a diamond embedded in rock, its value can only be realized through complex extraction processes.

In research and development circles, the Iron Triangle refers to three interconnected and interdependent constraints on bringing products to market. In Data Extraction, the same principle applies, but with different constraints. The three constraints of Data Extraction are:

1. Accuracy
2. Speed
3. Cost

Balancing these three constraints is the key to successfully achieving the goal of 99% accuracy without delivering the data too slowly or at too high a cost. The interaction of the three elements typically occurs in this manner:

• Aiming for higher speed can hurt levels of accuracy and can negatively impact costs
• Lowering costs can impact accuracy and speed
• Gaining accuracy usually slows speed and increases the cost

The only way to achieve the needed balance is through the use of smart technology.

Smart software, in combination with intelligent processes, not only delivers more accurate data extraction, it delivers it at a reduced cost and in a timely manner. But extraction is not enough, the data must also be structured and inserted into databases or structured data documents such as XML or JSON to be useful to other business processes.

LoanLogics created, and has utilized this approach from its inception, to create a proprietary process that approaches 99% accuracy and achieves industry changing turn times.

The Iron Triangle can help you achieve your goals or rust you out of business.

About the Author

David Gitlin

David Gitlin is LoanLogics Director of Technology. He is an industry-leading innovator with extensive expertise at the frontier where documents and data intersect. David is the author of a patented technology for integrating scanned text with Excel and has written and presented on numerous topics related to that field. As a member of the Management Team responsible for the initial launch and dramatic success of the Neat Company, David spearheaded the development team that created the revolutionary Neat Receipts platform. He and Neat Company were early pioneers of integrated image processing, OCR, text parsing and text recognition for consumer and business applications. At LoanLogics, David is leveraging his important earlier work by overseeing the development and refinement of the LoanHD® IDEA™ platform, the next generation document automation tool that integrates the latest in data extraction strategies and technologies. LoanHD® IDEA™ combines sophisticated workflow and reporting capabilities to create processes for data extraction that are flexible, efficient and highly accurate. David has an MS in Computer Science and an MA in Political Philosophy.