I was consulting for an oil and gas service company when they decided to add a new service line for their drilling company clients. I'll explain the system in plain English first, because the mistake only makes sense if you know why it mattered.
When an oil company wants to understand a formation, they cut a core sample out of the rock. The old way works like this: you core a stretch, then pull every length of pipe out of the hole to get the sample, then run all of it back in and do it again. On a deep well, one round trip can take 12 hours.
The new system let us leave the pipe in the hole. We'd bring the core up the inside of the pipe on a cable, send the empty barrel back down, and cut the next section. Same job in 30 to 45 minutes.
That only works if the inside of the pipe is right. The core barrel travels up and down inside the drill string, so the inside diameter (ID) had to be strictly between 4.125 inches and 4.187 inches. That's less than a sixteenth of an inch of room.
We found a manufacturer who said they could hit those numbers. I stayed on them about it, went back to the tolerances again and again, and every time they told me it wasn't a problem. So we ordered the drill string, 1.4 million dollars worth. Then we ordered everything else that had to go with it: another half million, bringing the total to 1.9 million dollars.
The company sold its biggest, most loyal client, a big exploration company, on running an entire winter project with the new system, 25 holes. The exploration company had hired a drilling company to run the rig, and my client was brought in as the service company supplying the specialized string and tools.
Here's my mistake, plain and simple: I never went and checked the work myself. I took the manufacturer's word every time instead of putting eyes on the actual pipe. It wasn't my money, but it was my job to make sure that order was right, and I didn't.
A week before the project started, the drill string showed up at a staging yard near the job site, and my gut told me something was off. It wasn't a thought. It was a heavy feeling I couldn't shake, like something had been missed and it was serious. Looking back, I think my stomach already knew what my head didn't: none of that pipe had ever been inspected.
Lucky for us, I had another consultant working near the yard. I asked him to go check the outside diameter and inside diameter of the string himself. Took him all day.
What he found was a gut punch. The outside diameter needed to be at least 5 inches, and it was, 5 to 5.25 inches across the board. But the inside diameter had to land between 4.125 and 4.187 inches, and it wasn't even close. The biggest one he measured was 3.75. Some were as small as 3.5.
The whole system was dead. 1.9 million dollars of steel and components we couldn't use, on a project starting in a week, for their biggest client.
That night I sat in the yard office at 10 pm, shocked and mad at myself, quietly doubting everything. Had I made the mistake? I went back through every email and every fax between me and the manufacturer. I replayed every conversation in my head. Every single one had the right tolerances. I'd sent them everything. The only thing I hadn't done was put my hands on a single piece of that pipe.
The client had to be told. But I wasn't going to call without a plan, because I knew the exploration company was counting on that string and our tools to drill the whole well. So before I picked up the phone, I lined up an older system the company already had, one the exploration company had never run before but that I knew worked, and confirmed the standard drill pipe for it was available through the drilling company they'd hired.
When I told the exploration company the new string was out, that none of it could be used, they went straight into panic mode. The drilling company had already started moving the rig onto location and had no pipe for our system to drill with. Once they calmed down enough to listen, I walked them through the plan: we'd run the older system on the drilling company's standard drill pipe, and it would still save real time on trips out of the hole, just not as much as the system that was supposed to be there.
We looped in the drilling company to line up the pipe. The exploration company agreed to run the first hole with the older system and said they'd decide from there whether it was worth staying with us.
They stayed. My client kept its biggest customer.
The pipe worked out better than I deserved. It hadn't been heavily modified and it fell well within standard safety specs, so the manufacturer and the company were able to sell all of it to other companies. They broke even. I'd love to say that was planning. It was luck.
What I'd do differently is simple. Before ordering the full run, I'd have had the manufacturer build a small batch of 5 or 10 pieces, and I'd have inspected every one myself. Then I'd have made the 150 kilometer drive to their shop at least once a month while the rest were being made. Anything out of spec, mark it, send it back, don't let it leave the building. I had put every tolerance in writing, and it didn't matter. Writing it down is not the same as checking it. The only reason my client kept its customer is that I had a backup plan ready before I ever had to admit the first one had failed.
Here's what I'd want any owner or consultant to take from this. It doesn't matter what the project is or what's being built. If it matters to the business, make the time to go see it in person. "They said it's not a problem" is not the same as checking it yourself. Every time.
