In our last blog “The UMPTeenth reason for agile data management,” we noted that upstream companies are using digital technologies to accelerate their upstream market pull transformations (UMPT). But we never said it was easy.
In recent discussions with a client organization’s analytics team, we got a first-hand account of some of the challenges keeping them from getting to the next level of data and analytics transformation. We utilized an empathy map exercise to gain deeper insights. We found the following:
- What they say – the speed of transformation is slowed by their inability to scale.
- What they think – we want to change, but we’re having difficulty moving the needle.
- What they see – promoting the development and consumption of data products, enhancing end-user adoption and tying results back to strategic themes are all challenging.
This team, like so many others we’ve worked with, wanted to build an agile, responsive analytics capability. But they were seemingly stymied at every turn. The good news is that none of the obstacles they face are insurmountable. Here’s how they can overcome them to build analytics agility at scale.
Balance the scales
To truly scale analytics across an enterprise and support the UMPT, a significant data transformation overhaul is needed. The following model sets out the key elements that, when executed successfully, facilitate the required data transformation. We have used this model with several clients and seen how powerful it can be.
<<< Start >>>
<<< End >>>
Though the model presents the key components of a digitally-enabled analytics transformation, it doesn’t lay out a sequence of actions. For example, an organization’s journey to cloud (depicted at the bottom of the above graphic) is a key enabler of an analytics transformation and should be the starting point. The reason is that achieving analytics at scale and with the necessary agility requires capabilities in cloud integration, continuous deployment into production environments, on-demand release management and telemetry (or receiving data back from deployed or released solutions). By focusing on the journey to cloud first, organizations will find that “regret work” is minimized.
As organizations advance their journeys to cloud, we believe they should also start scaling the analytics side of the model (depicted by the top right three squares in the graphic). This will allow their data science managers to shift their focus from the development of ad hoc proof of concepts and prototypes to streamlined analytics factories. Importantly, an early focus on the organizational model, adoption plan and potential value of analytics will help the organization define and implement the core data management elements (the top left three squares) that will be needed to underpin the transformation.
Choose the path for action and growth
So, how can organizations ensure that their analytics capabilities are maturing at pace?
To begin with, companies need to create a guiding coalition. This could be something like an analytics center of excellence (CoE) residing between the business and IT function. While there are pros and cons to using CoEs, we’ve found that one of the following three models is usually effective depending on a company’s current situation:
<<< Start >>>
<<< End >>>
After setting up the CoE, analytics teams need to establish a standard CoE operating model that will be adopted across the enterprise. This operating model will differ from one organization to another, but it should enable the following steps: intake, triage, frame, experiment, prepare, commercialize and operate.
We’ve recently set up an analytics CoE at a client site. As part of that effort, we interviewed the analytics teams across the organization (i.e., at the central IT, business unit and functional levels) and identified commonalities across existing processes. From that information, we created a “one-size-fits-all” operating model for tracking analytics products throughout their lifecycle.
With the new operating model in place, we had a consistent language across the company and could start tracking analytics product ideas. Importantly, the new operating model also helped us define the data requirements, more easily identify the critical data elements underlying each use case and work with the data organization to create a standard roadmap for data ingestion, cleaning and provisioning.
Next? It’s time for quick wins. Selectivity is key at this stage. With the right use case(s), analytics teams can quickly wow their colleagues, garner interest in the program and make their case for additional funding. But what is the “right” use case? The one that creates quick and notable value. For example:
- In upstream environments, companies may want the entire production organization—from the executive team to the lease operator in the field—to be on the same page in terms of daily production, field asset downtimes and speed of issue identification and resolution. In that scenario, the company might consider an enterprise production analytics hub. The hub could provide real-time information and actionable insights from well-to-basin level.
- In downstream environments, companies may be looking to maximize margins during the sale of products. To do so, they need to be plugged into bargain procuring opportunities, asset health and production availability, and trading opportunities to sell the right product mix at the right time on a near real-time basis. Choosing a use case that touches users across multiple business units helps make a case for analytics at scale, since the solution will deliver value to multiple user personas.
This leads us to the next focus area: adoption.
Attain critical mass
Every technology solution goes through an adoption lifecycle. You know the players…
<<< Start >>>
They recognize a problem, conceptualize a solution and are off to the races.
<<< End >>>
<<< Start >>>
The Early Adopters
They see the value in the solution and adopt without questioning.
The Late Majority
They balance risk and reward by playing catchup, hoping to be early enough.
<<< End >>>
What are some ways to cross this chasm and get the Late Majority to adopt the solution as early as possible? One tried-and-true method is for business and IT (with the help of the analytics team) to work hand in hand on a strategic initiative aimed to propel the enterprise forward. Having the CEO and leadership team drive such a top-down initiative is critical to successful adoption. This approach stands in contrast to “bottom up” approaches, where innovation starts within a single business unit. Though likely a good point solution for that group, establishing analytics at scale from such a customized and/or geographically dependent use case can be an uphill battle.
Irrespective of how an organization identifies the right use case, gaining critical mass requires constant adaptation. This means doing more than incorporating continuous improvements to optimize a solution. It means adapting the solution continually in response to how the insights are used. If there are plant/field users or executives who like to access, approve, review or revise data on the go with their smartphones, the solution must be mobile-enabled. If the team wants the convenience of daily consolidated insights or the ability to take actions from within their inbox, the solution team needs to make that functionality available. Customer-centricity is the key mindset shift to drive user adoption.
Capturing the value of agility
As energy companies throw millions (if not billions) of dollars at digital transformation, business leaders and shareholders alike are expecting dividends early on. Return on capital employed (ROCE) has morphed into return on digital employed (RODE). For organizations going through this process, the highest immediate return is seen through analytics.
The client we mentioned at the beginning of this piece realized more than $1 billion in analytics-driven business value in 2020 alone. Those returns are the result of a multi-year-long journey to scale analytics across the enterprise.
Others can reap similar benefits. But doing so will require companies to have the right data and analytics products in their solution portfolio to deliver value. It is also increasingly apparent that they will need Agile data management capabilities and the right scaling strategies to ensure they are maximizing their capital investments. These are the imperatives and key success factors for oil and gas companies moving into the 2020s.
Disclaimer: The views and opinions expressed in this document are meant to stimulate thought and discussion. As each business has unique requirements and objectives, these ideas should not be viewed as professional advice with respect to the business.