Miner earnings reports break a bitcoin mining company into two businesses: a power buyer and a coin seller. The revenue line shows bitcoin produced and sold. The cost lines show the electricity, hosting and depreciation behind each coin. Read those lines together and you can get a sense of whether a miner made money because it operates well or because the bitcoin price moved.
The most important qualification is that these are filing-based figures, not live dashboards. A quarterly report describes a period that ended weeks before publication, and every number in it carries that reporting date. Prices, difficulty and power costs can all move between the period end and the filing date. For related coverage, see How Macro Data Releases Move Crypto Prices.
This piece walks through the standard line items in a miner's quarterly report, what each one measures, and where readers get misled. It is information about filings, not investment advice; crypto markets are volatile and losses are possible. We covered a connected angle in How the Federal Funds Rate Moves Crypto Markets.
What does self-mining revenue actually represent?
Self-mining revenue is the value of bitcoin a company mined with machines it owns. The reported figure reflects the coins' value when they enter the company's books.
Two things sit behind that single number. The first is production: how many coins the fleet earned during the quarter. The second is realization: whether the company sold the coins or held them. A miner that holds everything can show thin revenue and a growing balance sheet at the same time. A miner that sells daily shows revenue that tracks the market almost in real time.
Check the filing's disclosure of coins mined versus coins sold. The gap between those two figures tells you whether the income statement is describing an operating business or a holding strategy.
Why is cost of revenues the line that matters most?
Cost of revenues is the direct cost of producing the coins: mostly electricity, plus hosting fees when machines sit at a third-party site, plus pool fees and direct site operating costs. Subtract it from self-mining revenue and you get mining gross profit, the closest thing the industry has to a margin figure.
Electricity dominates this line. That is why miners disclose power cost per bitcoin produced. The metric is simple in concept: total power spend divided by coins mined. When network difficulty rises, more energy goes into each coin and that per-coin cost climbs even if the power price never changes. When the bitcoin price falls, the same per-coin cost eats a larger share of revenue.
This is the mechanism to watch, not the headline margin. A miner can report a strong quarter purely because coin prices rose during it. The cost-per-coin line shows whether the underlying operation improved, held steady or got worse.
How do hosting segments differ from owned sites?
Many public miners run two segments. In the owned segment, the company buys power directly and bears the bill. In the hosting segment, the company's machines sit at a partner's facility and pay a hosting fee.
The filings separate these because the risk profile differs. An owned site carries the power price and the uptime risk on the company's own books. A hosted site converts some of that into a contracted fee. Readers can compare gross margin across segments to see which model performed better in the period.
The same split appears on the revenue side for companies that host machines for others. Hosting revenue is a service fee, not coin revenue, and it should not be read as mining income. Confusing the two is a common error in reading these reports.
What other line items deserve attention?
A few recurring items round out the picture:
- Depreciation. Mining hardware wears out and loses value fast. Depreciation is often a large non-cash cost, and it can make a profitable-looking quarter turn negative at the net income line.
- Impairments. When machine values fall below their carrying value, miners write them down. A large impairment signals that the fleet's economics deteriorated, often after a price drop or a hardware generation change.
- Hashrate disclosures. Filings state deployed hashrate, the computing power pointed at the network. Compare it to the prior period to see whether growth came from new machines or from acquisitions.
- Energy efficiency. Some filings report energy use per unit of computing power. Lower is better, and the figure shows how exposed the fleet is to older hardware.
- Derivative positions. Miners that hedge revenue with forwards or options disclose the positions and any gains or losses on them. These can swing reported results without changing a single machine's output.
Live profitability trackers show the same mechanics at retail scale. According to Minerstat's mining profitability data, profitability rankings price in the user's electricity rate alongside coin price, network hashrate and device efficiency, refreshed on a short cycle. The same inputs, at corporate scale, are what a miner's cost lines capture for a whole quarter.
What this means for reading the next report
Our analysis of the filing structure comes down to a short checklist. First, find the reporting period and treat every figure as of that date. Second, split self-mining revenue from hosting revenue. Third, pull out cost of revenues and compute or locate the per-coin power cost. Fourth, check whether mined coins were sold or held. Fifth, look at depreciation and impairments before trusting net income.
That sequence answers the question a headline usually obscures: did the operation improve, or did the market do the work? A quarter with flat production, flat power cost and a rising coin price is a market story. A quarter with falling per-coin cost at steady production is an operations story. The filings contain enough detail to tell the two apart, but only if each line is read against the date it describes.
What remains unknown from any single report is forward exposure. Power contracts, machine delivery schedules and hedging books change between filings, and the next quarter's cost structure may look little like the last one. The report describes the period it covers; the mechanism, not the moment, is what carries over.
