By now, most companies recognize that there are huge opportunities to use advanced analytics to improve decision-making and gain competitive advantage. Research shows that when companies embrace advanced analytics they can deliver profit gains that are 5% to 6% higher than those of their competition. One report even shows that companies enjoy an average benefit of $10.66 for every dollar spent on predictive analytics.

The oil and gas industry is ready for advanced analytics. In 2014, spending on domestic tight oil development will top $72 billion, with 30% growth by 2020. Yet despite all this money and attention, one in four wells performs significantly below expectations. Producers can’t afford to be this wrong when they are spending $1.50 for every $1 that they get back this year.

Why advanced analytics?

Today’s advanced analytics employ sophisticated data processing techniques to yield information from cross-functional datasets that are brought together from inside and outside the company.

There are thousands of multistage lateral wells with hundreds of data points per well in the mature shale plays, but only a subset have complete data and only for some of the parameters. Traditional analytics techniques can only measure correlations across complete datasets. This results in a lot of valuable data being ignored. Advanced analytics techniques extract much more information from these “sparse” datasets and measure correlations from the available data.

Predicting production in shale reservoirs

Predicting oil and gas production in unconventional reservoirs is an ideal application of advanced analytics. There is a huge amount of data available, and the industry has already proven that valuable insights can be derived.

Public information is available from organizations like FracFocus and state agencies such as the North Dakota Industrial Commission and the Texas Railroad Commission. This includes well location, design, production and geological information. Every operator also has lots of proprietary data about its wells and more detailed geological information.

Of course this isn’t news to the shale reservoir engineer who regularly uses these data to make critical decisions about where and how to drill expensive wells. But the inability to process all of this information combined with the inherent limitations of predicting average production for large acreage positions based upon a single well design forces engineers to make decisions using limited models.

Advanced analytics make it possible to instantly evaluate short-term production, EUR and a financial forecast for a specific well design at a specific drilling location. This allows reservoir engineers to evaluate far more well designs per well than what is possible making type curves with traditional tools.

Gaining new insights

Advanced analytics provide better accuracy and efficiency day to day. But the real value comes from the big strategic questions that operators struggle with, questions such as is there a single well design for a shale reservoir or should each well be designed individually to minimize dollars spent per barrel? Which completion design parameters are most important, and how do they relate to each other? What are the short- and long-term implications of pumping sand vs. ceramic proppant? How does this change when using multiple perforations per cluster? How are these factors impacted by different reservoirs?

Managed services approach

There are many moving parts to the typical advanced analytics project. These include collecting and preparing data, performing the analysis, creating ensembles of models, running simulations, generating reports and analysis, delivering interactive visualizations and then keeping the analysis evergreen by repeating the process as new data become available.

This approach requires months to years for even an experienced cross section of IT specialists, data scientists, user interface engineers and subject matter experts.

Rather than building all of this from scratch, OAG Analytics recommends going with a managed services approach. It automates the data extraction, transformation and load function. The models are then built and stored in a cloud infrastructure and made available 24/7. Prebuilt analysis and simulation are designed to help solve specific problems for the roles within an organization from the geoscientist to the reservoir engineer to senior management.

Preparing for advanced analytics project

Gather the data. Advanced analytics projects start with a detailed analysis of publicly available well data as well as all available geological and geophysical information. More comprehensive proprietary drilling, completion, cost and production data are then added to improve accuracy and provide more robust insights.

Although it may sound daunting to gather all this information, it’s easy enough to get started by collecting information to gain some initial insights and then adding more data as they become available. Working with an analytics partner who knows unconventional data and is equipped to do a lot of the heavy lifting will get the process going much faster.

Understand the desired outcomes. As Stephen Covey points out, companies should “begin with the end in mind.” What does a company hope to accomplish from the analytics initiative? Does it plan to incorporate the results into operating practices? Identify a set of goals and metrics that can be measured.

Test and continue testing. It is critically important to blindly test the predictive model against wells with known production. Continue to test as new wells are drilled. Rebuilding models from scratch as new data become available maintains and improves accuracy in volatile data environments such as unconventional reservoirs.

Don’t turn it into an IT project. Many vendors will want to sell a box of tools for a company to build its own advanced analytics solutions. Others provide analysis with no means to leverage the insights. Go with a managed-services provider that can stay on budget and rapidly deliver to the defined desired outcomes.

Analyze the results and look for areas to exploit and improve. An advanced analytics initiative is a journey. Initial results should enable a company to start asking questions that it didn’t know it could ask. This leads to new learning and the desire to explore. Experiment by working with an analytics partner that will enable the company to test many hypotheses.

Periodically refresh the data. Advanced analytics are extremely adept at filtering out real data from “noise.” But the analytics are only as good as the input data. As new information becomes available, rebuild the models. Automate the data refresh process as often as can produce measurably improved results.

Drive adoption through the organization. A company must define what it will take for it to trust the results. Once that’s done, it can evolve the current processes to leverage them. Competitive advantage can only be gained if the business adapts.

Roll up the value to other parts of the company. Keep in mind that advanced analytics can go well beyond well production prediction and can also be used to provide a unique competitive advantage for evaluating acquisitions and divestitures, forecasting, reserve reporting, simulating well spacing scenarios and more.

Get started now

Advanced analytics are here to stay. New insights that are changing shale reservoir development are constantly being discovered. More data become available all the time. Companies must start considering how advanced analytics can help the organization gain efficiency and accuracy from the data it already has.