Like oil, data is a commodity with tangible value. Yet the digital oilfield is like the wild west, where energy professionals spend an inordinate portion of their day on data wrangling because of a chaotic mix of structured and unstructured data, multiple versions of the truth, and lack of data governance. Every department in the energy enterprise is contending with increasing volumes, variety, and velocity of data, leading to analysis paralysis, delayed decision-making, and unnecessary rework (e.g., prior period adjustments). This “big data” dilemma often leaves important questions difficult to answer quickly, including
- Do geoscience, drilling, and completion teams have the most accurate data sets to optimize well placement and performance?
- Are mineral and royalty owners getting paid based on accurate volumes and interest decimals?
- Do field and production teams have clear situational awareness to manage assets effectively following a merger or acquisition?
- Is compliance providing the most accurate production and environmental reporting to state and federal agencies?
- Are land, production operations, accounting, and regulatory making decisions with the current/correct well status?
For organizations that do it well, data management provides a competitive edge in an increasingly digital oil field that accelerates business performance and sets every department up for success. But teams all too often are so busy managing all the moving parts of data management (not to mention the business of finding, extracting, and moving hydrocarbons) that they take their eye off of “the prize”—the payoff after you have put everything into place to sustain successful data management.
In terms of attaining the prize, there are many business goals data management should achieve for work flows across the energy enterprise, including land, drilling and completions, geoscience, production management, field operations, and even human resources. Think about the impact of poor data management on an company's cash register, from allocations and well tests to sales and financial statements. No matter the price of WTI or Henry Hub, even a little bad accounting data or rounding error can hurt the bottom line.
For example, inconsistent use of data standards introduces the possibility of missing an obligation to interest owners if the land department uses API-12 while the drilling department uses API-14, preventing accurate tracking of a wellbore trajectory or new sidetrack on a lease. Or consider a midstream example where fixed asset accounting depends on high-quality location data for an interstate pipeline. Lat/long for preliminary right of way, permitted, and as-built designs can vary, and, with hundreds of miles to account for across dozens of city, county, and state lines, teams struggle to stay compliant. These examples share a similar risk where even a small discrepancy to asset location can have massive financial implications, underpayment to interest owners in the first and underpayment to state and municipal taxing authorities in the latter, all hinging on effective data management.
Then there are new prizes (or burdens, depending on how you look at it) as regulatory and compliance obligations expand. Companies must manage an increasingly complex variety of environmental data for accurate HSE reporting (e.g., audio, visual, olfactory inspection of oilfield assets). The “E” in ESG is only making the need for effective environmental data management more urgent as greenhouse-gas reporting increases in scope and complexity. As part of the recent Inflation Reduction Act, oil and gas companies now face a methane fee of $900 per metric ton starting in 2024 (increasing to $1,500 through 2026), making it absolutely imperative to master regulatory data management to avoid over- or underpaying the government.
Prizes of the future include predictive analytics (PA), machine learning (ML), and artificial intelligence (AI). The oil and gas industry has already seen some success in areas like artificial lift optimization and drilling automation, but many PA, ML, and AI initiatives stall out or plateau because they need big data to thrive.