These days it seems every energy-related trade journal, conference and company touts the potential of “Big Data”.  Proponents all run the same BD examples up the flagpole - GE’s jet engine database, Amazon’s analysis of customer buying habits, and the Internet of Things – and say that BD will change the oilfield by improving efficiency and lowering finding and development costs. Well. While someday the oilpatch may reach the Promised Land, we don’t think it’s going to happen anytime soon. 

First, where BD can most help the upstream oil and gas industry is far too late in the investment cycle to meaningfully reduce costs. If you read thought pieces on how the oilfield could benefit from the adoption of BD, the discussion focuses on production. In these examples, BD proposes to improve operating efficiencies by looking at the millions of wells producing oil and gas because THIS is what looks most like other BD application such as GE’s jet engines and Amazon’s customers. 

However, if well production is our focus, we’re too late to move the cost needle since typically 80-90% of a field’s total spending takes place up front in lease acquisition, well construction, well completion and the initial production set up. And the issue in the production sector will be whether operational savings that might reach 15-20% at the maximum are enough to offset the cost of gathering and analyzing data. 

Secondly, if BD’s bright light focused earlier in the spend cycle on drilling or completion we think there are limits that would appear to barricade much of the upstream oil and gas industry from adopting, much less benefitting from, BD.  Why? Call it “tribal knowledge”.

Clearly there are some companies with large-scale drilling programs that may benefit from a BD approach, but the fact of the matter is that almost all projects executed anywhere in the world – but particularly in the US on land – are led by small teams of smart people using small or regional service companies to drill and complete a limited (< 100) number of wells.  These small teams develop the “tribal knowledge” for their very specific projects…tribal knowledge learned through trial and error that is specific to their field, their rock, and their available service providers. 

The lessons learned from drilling and completing the Wattenberg cannot be applied to the Vaca Muerta, although the general process is the same. The SCOOP in Oklahoma fracs different than the STACK play just a few counties over. Big Data only works when tribal knowledge is a limiter, not an enabler. And it would appear that almost every drilling and completion project around the world benefits far more from a tribal knowledge “micro” approach than a BD “macro” approach.

Finally, the other issue blocking the application of BD to the oilpatch involves ownership of the subsurface data. Even if Wattenberg lessons could be applied to the Vaca Muerta, if the operator who owns the Wattenberg data doesn’t make it available to the Vaca Muerta operator then a BD approach will not bear much fruit. The industry will have to figure out how to value data – a process that has yet to begin - before it can fully realize the promise of BD.

A production-first focus, tribal knowledge, and data valuation – three factors which will slow the use of Big Data in the oilpatch.