Now, more than ever, oil and gas supply management organizations have an opportunity to generate value for their enterprises beyond rate reduction negotiations with suppliers. They sit on a vast ocean of enterprise data including demand, prices, contracts, emails and production output. Moreover, a virtually infinite amount of external online data is at their fingertips and waiting to be leveraged. These factors help enable opportunities to improve cost structures and risk management, identify revenue-generating opportunities, increase automation, and establish desirable working relationships with vendors and suppliers.
Machine learning is a driving force behind this unprecedented opportunity. Here, algorithms learn from vast amounts of data and make predictions without being explicitly programmed. Despite the technology and data being readily available and accessible, our experience indicates that many organizations still fall short of capturing the potential value.
Why is it so difficult?
Advanced analytics requires a shift in mindset. Many supply management organizations jump straight into developing predictive models and algorithms without a clear understanding of the drivers, sequence, and requirements for a satisfactory proof of concept, pilot, and deployment. This creates an outcome with little promise of value and erodes the credibility and trust of the supply management organization across the enterprise.
Many predictive analytics vendors provide self-serving platforms that require little understanding or experience of machine learning, data science, or even basic statistics to develop predictive and prescriptive models. However, the organizations that successfully implement advanced analytics resist the urge to jump right into algorithms and technology, instead taking a systematic approach.
People and process first, then technologyAdvanced analytics effectiveness starts with people. The primary drivers that help increase the rate of progress for any supply management analytics initiative are:
People: Start with the question, “What are we solving for?” and ensure the answer aligns with the corporate strategy. A definitive answer and support from leadership are leading predictors of effectiveness. Such initiatives start with the CEO (ask Jeff Bezos), CFO, CIO, Chief Supply Chain Officer (CSCO), and Chief Procurement Officer (CPO). Establish a clear understanding of leadership’s buy-in and commitment, followed by a clear understanding of the organization’s analytical talent and how to deploy it. Is there a balance between information technology, data science, and business analysts?
Process: Good data is king, and data governance is the vehicle to help you achieve alignment and ownership, assign responsibilities, establish decision-making processes, determine escalation points, and break down data silos. From the outset, understand the data quality, breadth, level of integration, and accessibility. Then establish the scale of your company’s analytical resources and how to manage and approach them from an enterprise-wide perspective.
Technology: Assess the technology required, with a clear understanding of the outcome. Maintain a holistic view that accounts for data ingestion and preparation, data storage (such as data lakes) and retrieval, data usage (including analytics algorithms and models that leverage cloud computing architectures), and data visualization.
Changing environments require new approaches. The path to an effective advanced supply analytics journey starts with a clear vision of the end state, enabled by the right organizational structure and facilitated by people, processes, and technology. Today, oil and gas supply management organizations live in a data-rich world that presents an unprecedented opportunity to create economic value.