Oil and Gas Analytics

Luther Birdzell spent 20 years building software to make data more valuable to subject matter experts. After exiting his previous company in 2013, he saw a clear opportunity: AI and machine learning were becoming the most transformative technologies for unlocking value from existing data, and the oil and gas industry—which was the fastest-growing sector of the US economy at the time—was desperate for solutions. He zeroed in on the most capital-intensive problem: well-planning optimization. Every year, $500 billion is spent upstream in oil and gas, with 90% of that ($450 billion) spent in the 30 days before and after drilling a well. This was the highest-leverage place to apply machine learning.

Luther bootstrapped R&D starting in 2013, spending two years building the technology before going to market. The initial vision was cloud-based data management, advanced analytics, and insight visualization purpose-built for geologists, geophysicists, and petroleum engineers. The key differentiator: no coding required. Engineers could build machine learning models in less than five minutes, and crucially, they could challenge the computer's analysis through the lens of physics and domain experience—something critical in high-consequence industries. The platform moved data management from days to minutes.

In Q1 2017, Luther launched his go-to-market. By mid-2017, he had signed his first three customers. From the start, Luther understood this wasn't a self-serve SaaS play. He built a high-touch enterprise model: his consultants (all with geology, geophysics, or petroleum engineering degrees, many with PhDs) worked directly with oil company engineers on a peer-to-peer basis. This proved critical for adoption. The initial customer acquisition cost ranged from $5K for a local Houston customer to as high as $50K for out-of-state deals, mostly covering sales, technical pre-sales support, and travel.

By early 2018, the model was clearly working. Luther had grown to 5-10 customers paying $10-50K per month, generating "north of 50 grand per month in revenue." The unit economics were strong—each well saved $400K to $1M in costs, making the ROI obvious to buyers. One concern: churn. Luther was transparent that he hadn't yet had a full renewal cycle, but all customers on annual contracts were still active and renewals looked "very good." The team had grown to 20 people, with 6 dedicated consultants. Importantly, Luther had tested the sales model with one salesperson and proven it could scale, prompting the search for a Chief Sales Officer. For 2018, the company was on track to grow 300% year-over-year.

Headed into late 2018, Luther had raised less than $3 million from angels over five years of operation and was preparing to raise Series A in Q4. The capital would accelerate sales hiring and consultant hiring to support revenue growth. He'd proven product-market fit, built a solid team spread across Houston and remote locations, and had a replicable enterprise sales model. The opportunity ahead was clear: scale the model across the remaining 90% of oil companies he wasn't yet serving. At 41, married, and running a profitable, growing company in one of the world's largest industries, Luther had finally taken the startup plunge he'd been contemplating since age 20.