The compression market no one else can see. Unit by unit.
Every basin's installed horsepower, owned versus contract, assigned vendor by vendor, refreshed quarterly. The census no analyst, no vendor, and no operator has today. Then we measure every other OPEX line against its floor.
The denominator does not exist
Ask the public record for any basin's installed compression horsepower and the answer, across all 12 major US onshore basins, is undisclosed. Not thin. Nonexistent. Our census counts what never gets totaled: roughly 48M horsepower of operator-owned compression, about 74% of the installed base, invisible to every analyst, every vendor strategy team, and every operator benchmarking itself.
Scope: permitted field and gathering compression, ≥200 HP, excluding processing-plant and transmission compression, reconciled bottom-up where public disclosure allows.
can see today
Every unit, one schema, four times a year
The census is the installed base, normalized: unit-level rows across 12 basins, each carrying installed horsepower, owned versus contract status, an assigned vendor, and a confidence score trained on confirmed field-verified units. The refresh is quarterly, with adds and removals tracked, because the installed base moves monthly and a number that holds still is a file, not a product.
One census. Three ways to read it.
The market you compete in, unit by unit.
Your competitors report incompatible buckets; the census puts the whole board on one schema, with the recontracting windows nobody else can time.
Request a basin sample →The denominator the whole sector is guessing at.
Per-basin installed horsepower, scope stated, reconciled bottom-up. Citable in a footnote.
Get The Iron Count →You cannot prove your LOE is tight.
Peers tell you your rank, never your floor. We sell the floor, starting with compression.
Talk to The OPEX Desk →One cost line moves both levers of LOE
Should-cost is the floor: the theoretical bottom of an operating line, built up from the vendor's own economics. The gap between the floor and what you pay is recoverable, and no one else sells that number. We start with compression because it is the one line that pulls both levers of LOE per BOE: a large cost line in the numerator, and the biggest driver of runtime, and therefore production, in the denominator. Improve compression and the ratio falls even at flat cost. (You did not spend less; you produced more. The ratio cannot tell the difference, which is exactly the point.)
This is the RSMeans move for oil and gas operating cost. Publish the reference the market measures itself against, then defend every line to it.
See the method →One census, sold the same way to everyone who reads this market. Same rows, same price, published scope. No exclusivity, no early access, no side deals. That is why an analyst can cite it and a vendor can trust it.
"We saw both bottom-line cost improvements and runtime stability gains, which translated into more predictable production and stronger cash flows."
The roster the newsletter pulled in on its own: compression providers you would recognize, and one of the largest banks on the Street, all inbound. The advisory names behind the value figures: Jonah Energy and Anschutz.
Five rungs, one census
Paid data tiers are request pricing at launch. Contributors can lower their price by putting field data into the census through Ground Truth.
Start with the number
The Iron Count is free in full, one basin every Thursday, scope-stated. Nominally Hedged is where we show our work. If either earns a footnote in your next model, you will know whether the rows underneath are worth a conversation.