In 2013, I began working as a data analyst for Ruths Analytics and Innovations, a startup data science company in Houston, focusing primarily on Oil and Gas clients. I was looking for something different, and it was, considering I had been previously developing business intelligence solutions for federal agencies in DC.
For a young professional, it’s been a privilege to watch the rising popularity of data science in this industry and I’ve put together a few things I’ve noticed:
The role “Data Scientist” inside E&P remains cloudy
Who is a data scientist anyway? That is a question the oil industry has been trying to answer on its own. The seeds of the discipline are there, with talent being peppered through organizations. But like other sectors that recognize the tactical need for the data science skill, E&P companies have yet to officially create titles, departments, and organizations to support this function. Data science will continue to play a part in operations and field/well design; however, we don’t see the training, competency-building, or tech stack being deployed and developed in a systematic way to bring the science to the decision-making table.
Analytics is no longer a luxury
Before the oil price crashed, advanced analytics was more or less treated like an R&D project. When the industry began to cull their organizations, IT departments were asked to validate the need for many of their projects. R&D has become more of a luxury in this market environment. Those analytics projects that could articulate a value to the organization now were moved into the asset teams and put into production.
An industry that is looking for efficiency is going to be hiring a new kind of thought worker. They’ll be familiar with the latest professional technology and have domain expertise. This is what David Feineman, BP’s senior business process advisor called a nontraditional upstream discipline. Looking forward, this is a role that will need to be refined and clarified. Bringing a new role into an industry will require consistency, planning, development, and real dedication to the role of the data scientist in E&P.
Data foundations remain a nascent pursuit in E&P
When I left federal government work for oil and gas, I noticed a remarkable difference in how data is managed. In federal government, I saw strong use-cases for a data foundation as the consolidation of repositories were meant as a cost saving measure; reducing IT maintenance costs on the general ledger. Additional value was gained in this effort as data foundations create opportunities for business intelligence. I believe a key business driver for a consolidated data foundation comes from the federal government’s need to reduce redundancies.
In our case, E&P is experiencing its own budgetary crisis today. Where energy companies have previously fielded multiple repositories, the need for streamlined data management is now greater with cost saving initiatives being levied at IT departments. The quest for master data management remains elusive. E&P Company data terrain remains spread across legacy applications (read WellView, ProSource, OSIsoft PI) and traditional organizational structures like drilling, production, and geoscience.
So while data-driven analytics appears to be gaining recognition within E&P as a value-add, the data foundations and functional roles required to build, discover and leverage analytics remain undeveloped. It will be up to data scientists to continue to champion value-add through the application of provident analytics, while organizations will be best served if they embrace function-based roles for data science while continuing to build cross-organizational data foundations.
Technical Director at Ruths.ai