The Back Story
At Oseberg, we’re in the business of extracting oil & gas ownership data, state regulatory drilling information, legal provisions, legal locations, and much more from federal, state, and county public sources all over OK, TX and NM. The result is the most comprehensive commercial market intelligence dataset for those states in the industry.
In the course of scaling our operations and reducing our costs, we got very good at it. We developed our text annotation business process software because nothing in the market could do what we needed to process oil & gas information at the scale we required. We embraced artificial intelligence in the form of machine learning & natural language processing not to replace humans, but to make their jobs less tedious and to convert data entry folks to data analysts and quality control experts. We learned the balance of risks between machine text extraction predictions and human ones and continuously refined that process every day with thousands of documents.
Insert Your Challenge Here
Then customers started asking us if we could bring this data transformation process & expertise to their in-house data. So we started doing that also, quietly, to make sure we could be as good at that as we are with our public source data. We are.
We are now offering this data extraction technology, methodology & data quality engine to you for your data. Our Oseberg solutions practice can help you unlock the value of your in-house data to:
- Identify and enable new transformational data & document management practices and platforms.
- Prepare for & mitigate transactional buy-side and or sell-side due diligence risks well before a data room is ever created.
- Identify and mitigate contractual risks associated with e.g., lease provision drilling obligations, shut-in provisions, or midstream acreage dedications.
- Protect from unwanted litigation by hunting down specific legal phrases or word choices that are determining major court cases involving e.g., habendum clauses
- Exploit and/or protect horizontal Pugh severances by integrating & standardizing your depth and geologic formation references across completion, production, and lease provisions.
Your personal data creation factory
To scale our commercial data creation efforts, Oseberg built a supervised machine learning text extraction platform “dStill”. Across our business we use this platform to turn tens of thousands of document images into clean, standardized structured data that can be mapped, queried, and interrogated on a weekly basis.
Using dStill we have reduced our own lease processing time from initially around 30 minutes per lease using no workflow automation to around 10 minutes per lease document using image text extraction to now less than 5 minutes using AI/ML predictive algorithms and human QC. That impact has allowed us to reduce our internal data extraction costs by over 60%!
Now, through our Solutions team, we turn part of this factory over to you, enabling your large scale data creation efforts to proceed in record time and at an unprecedent low cost basis.