The unconventional revolution forced operators to ascend a steep learning curve to understand how best to access hydrocarbons residing in nano-sized pores. Reservoir characterization, drilling strategies, hydraulic fracture patterns and production schemes all have undergone intense scrutiny and, as a result, improvement.

One of the things that has remained constant throughout this metamorphosis is the industry’s absolute thirst for data, or at least data that lead to better decisions. But making sense of the increasing amount of data streaming into the industry every day can be overwhelming.

“The biggest issue we have in the oil and gas industry is that you can measure anything,” said Allen Gilmer, co-founder and executive chairman of DrillingInfo. “There are zillions of bits of information that you can collect, but it means nothing unless you can structure it in such a way as to query it for things that are meaningful.

“The devil is in that detail.”

Secret sauce

Gilmer said one of the company’s goals is to allow companies to test their models against what is generally known to quantify their competitive advantage in their understanding.

“When people are doing really interesting, groundbreaking stuff, we give them the tools necessary to quantify how much more they understand and create value with that understanding,” he said. “They can quantifiably realize what their secret sauce is worth. Every one of these companies has what they consider a secret sauce, some of which is spectacularly good and some of which is not so good. How do they compare? How do they know how they compare to what our general Drillinginfo membership already has available? We created a process to structure that system so that these random recipes can be evaluated against one another rather than just randomly floating about.”

Over the years Gilmer’s team has attempted to quantify the complexity of the oil and gas industry, an industry that he characterized as “having a lot of moving parts.”

“These pieces fit together, often in some very inelegant ways,” he said. “Trying to figure out what questions to ask and how to structure data that will make it meaningful is a challenge. There’s historical data, there’s real-time data, and then you have the big mystery of geology… and geology matters.”

This is a difficult concept to sell to shale players, whom he characterized as believing the unconventional plays are “post-geology.” “How these rocks behave, what they contain and how they react to stimulus is critically important,” he said. “Rocks are not uniform. They never have been.”

But the size of the resource is enough to get the company’s attention. Gilmer said the focus of much of its recent software development has been on structuring data such as massively automating the digitization and quality control of well logs. To that end, the company now has a machine-learning methodology for auto-picking sequence stratigraphy enabled by pattern recognition technology used outside the oil business.

“People used to go out, put their wells in their software and do their picks well by well,” he said. “A great interpreter could interpret 100 wells a day, and with our system he or she can interpret 50 wells in detail and then machine-interpret the 50,000 wells in between. It removes the rote aspect from the business.

“The dollars being paid to technical professionals should go to creative work that adds value to the company.”

Applying the latest digital technologies to these age-old problems has given Gilmer a new appreciation for the guys with the maps and the colored pencils. “You learn how good the old geoscientists were,” he said. “Talk about completely undetermined systems! When I had my old seismic company, we went to South Texas because we assumed they had left some stuff updip. I lost a lot of money because those guys were so good at what they did.”

All about automation

That being said, Gilmer feels that younger people are more willing to allow some of their workflows to benefit from automation. His company offers tools that automate geological interpretation, drilling activity and production forecasting, and he said it’s often harder for the older generation to accept the new reality.

“It’s just a way of thinking,” he said. “Young people have a lot more openness to thinking in these deterministic ways vs. those of us who came up thinking that you had to build everything from first principles and look for data to support your ideas rather than looking at data for ideas.”

He gave an example of a geophysicist client who pulled available WITSML data into the Drillinginfo Transform platform and analyzed it. He concluded the drilling engineers were building curve too fast and were about to get stuck. They told him to stick with geosteering and geophysics. Then they got stuck.

“He predicted it, and the management team told him and the drilling team, ‘Feed that to the drilling engineers on every well we’re drilling.’ They then told the drilling engineers, ‘You’d better listen to him. If you get stuck again, it’s coming out of your bonuses.’”

Part of the trick has been working around the preconceived notions that come with the traditional siloed approach to finding, drilling for and developing oil and gas reserves. Gilmer said, for instance, that no geologist ever wants to see anyone else’s picks in an interpretation.

“Our pick might not be the right pick, but we do a very complete breakdown as to why we made the sequence picks that we did, and we provide that to our clients,” he said. “I talk to geologists who say, ‘I don’t trust anyone else’s picks.’ And I say, ‘How would you like to see the whole basin architecture in the next five minutes?’ Well, of course they’d like to see that.”

And seeing it, in turn, helps them revise their picks without having to pick from the bottom up, he said.

The company also has developed new methodologies for determining EURs and quantifying the variability of those numbers. “When you build a forward production curve, the EUR that curve defines isn’t one number,” he said. “There’s a probability cloud … an uncertainty around that.”

Like the geologist and his picks, engineers are partial to their own EUR determinations. But when they’re able to see the big picture, including the geographic locations of their factors and shape curves by operator, they’re more aware of the factors that create the potential impact on the bottom line, he added.

Digitalizing the hunch

Having been an oil man, Gilmer knows the types of information operators want to know and is helping them obtain it through data rather than through gut feeling. “These are the kinds of things that you do from a very subjective view as an operator,” he said. “The point of view we had was to move it from the subjective to the objective, to create numerical models out of these things and to create a system that allows us to act on those models.”

Overall, Gilmer finds the encroachment of the digital era into the oil patch to be a fascinating development. “I think it’s an exciting time,” he said. “This is the most meaningful time in the history of the oil and gas industry. We think of the ‘golden days’ of maybe the ’40s and ’50s with the big majors making their plays. But I think it’s much more significant today.”

Contact the author at