What Is Bitcoin Mining and How Does It Work?

Bitcoin mining is the process of using specialized computers to confirm transactions on the Bitcoin network and add them to the blockchain, with successful miners rewarded in newly created bitcoin plus transaction fees. In one week of late August 2026, miners worldwide collected roughly 3,214 BTC this way, worth about $253 million. For a large share of those operators, the sum still barely clears the electricity bill.

That gap between headline rewards and hard costs is what makes mining the most misunderstood part of Bitcoin. This guide covers how bitcoin mining works, why the network depends on it, what it costs in 2026, and how to start if you decide it suits you.

Manage your Bitcoin securely with the self-custody Bitcoin.com Wallet app.

Key Takeaways

• Bitcoin mining confirms transactions and issues new coins. It is how the network stays secure without a bank, company, or government in charge.
• Miners who add a block currently earn 3.125 BTC plus transaction fees, worth roughly $245,000 at September 2026 prices. The reward halves again in 2028.
• More than 20 million of the 21 million bitcoin have already been mined, about 95.6% of the total supply. The rest trickles out until roughly 2140.
• Competitive mining requires an ASIC, an Application-Specific Integrated Circuit built to do one job: compute SHA-256 hashes as efficiently as possible.
• Profitability comes down to two numbers: your electricity price per kilowatt-hour and your hardware efficiency in joules per terahash. Cheap power decides everything.
• Mining on a phone or an ordinary laptop does not work. Apps promising it are simulations, ad harvesters, or scams.

What Is Bitcoin Mining?

Bitcoin mining is the work of validating Bitcoin transactions, bundling them into blocks, and adding those blocks to the public ledger known as the blockchain. Miners who do this work are paid in bitcoin, which is also the only way new coins enter circulation.

Conventional currencies are issued by central banks that can create more money when policy demands it. Bitcoin has no issuer. The rules Satoshi Nakamoto set out in the 2008 Bitcoin white paper cap the supply at 21 million coins and release them on a fixed schedule through mining alone. The name is a deliberate nod to gold: just as gold miners spend energy and equipment to pull scarce metal out of the ground, bitcoin miners spend electricity and computation to earn scarce digital coins. Nobody receives new bitcoin for free.

Satoshi Nakamoto posting on Bitcointalk in 2009, when mining still ran on ordinary CPUs. The shift to GPUs and then ASICs arrived within two years. Source: Bitcointalk.

Mining has changed beyond recognition since the first block in January 2009. Early participants mined on laptop CPUs and earned 50 BTC per block. Miners moved to graphics cards, then to purpose-built ASIC machines, and the industry today is dominated by warehouse-scale operations running on industrial power contracts. Cambridge researchers found the hardware market is now strikingly concentrated, with a single manufacturer holding 82% of ASIC market share and the top three controlling more than 99%. The underlying principle has not moved: whoever contributes verifiable computational work to secure the network gets paid by the network.

How Does Bitcoin Mining Work?

Every ten minutes or so, miners around the world race to solve the same puzzle. One wins, adds the next block of transactions to the blockchain, and collects the reward. The rest discard their work and immediately start again. That cycle has repeated more than 965,000 times since January 2009, and it breaks down into five stages.

Transactions Wait in the Mempool

Unconfirmed transactions waiting in the mempool on September 11, 2026. Miners select from this queue when building the next block, generally favoring higher fees. Source: Bitcoin.com Explorer.

When someone sends bitcoin, the transaction is broadcast to the network's nodes and lands in a waiting area called the mempool. Every miner watches this pool of unconfirmed transactions, and each one carries a fee its sender attached as an incentive for inclusion.

Miners Build a Candidate Block

Each miner assembles pending transactions into a candidate block, generally prioritizing those paying the highest fees per byte of data. The miner also adds a special entry called the coinbase transaction, which pays the block reward to their own address if they win.

The Hashing Race Begins

The header of block 966,465. The nonce of 2,357,047,714 is the number miners changed repeatedly until the block hash fell below the target set by the difficulty of 127.45 trillion. Source: Bitcoin.com Explorer.

Now the competition starts. Miners run their candidate block header through SHA-256, a cryptographic hash function that converts any input into a fixed-length string of characters. The network accepts a block only if its hash falls below a specific target value. Because hash outputs are unpredictable, the only route to a winning hash is trial and error: the machine changes a small number in the block called the nonce, hashes again, and repeats quadrillions of times per second.

Picture a lottery where every hash is a ticket. You cannot make any single ticket luckier, but you can buy tickets faster than your competitors. The combined speed of every miner on earth is the network hashrate, which stood near 934 exahashes per second in early September 2026, with one daily reading on September 4 touching roughly 1,001 EH/s. At that pace the network computes close to a quintillion hashes every second.

The Winner Broadcasts and Gets Paid

Block 966,465, mined September 11, 2026. The first transaction pays 3.13236197 BTC to the winning miner, the 3.125 BTC subsidy plus 0.00736197 BTC in fees from the block's other 6,777 transactions. Source: Bitcoin.com Explorer.

The first miner to find a valid hash broadcasts the block to the network. Other nodes verify it in a fraction of a second, since checking a solution is trivial while finding one is expensive, then append it to their copy of the blockchain and restart the race. The winner collects the block reward: 3.125 newly created BTC plus every transaction fee inside the block.

Difficulty Adjusts Every 2,016 Blocks

Ten consecutive blocks mined on September 11, 2026. Intervals ranged from roughly four to eleven minutes, which is why the ten minute target is an average rather than a schedule. Source: Bitcoin.com Explorer.

The protocol targets one new block roughly every ten minutes regardless of how much hardware joins or leaves. So every 2,016 blocks, about every two weeks, the network recalculates mining difficulty. Blocks arriving too fast push difficulty up; blocks arriving too slowly pull it down. On September 6, 2026, at block height 965,664, difficulty rose 1.31% to 127.45 trillion, the eighteenth adjustment of the year.

All of this together is the Proof of Work consensus mechanism. It ties ledger security to real-world energy expenditure, which makes rewriting bitcoin's transaction history prohibitively expensive. An attacker would have to redo the accumulated work of the entire honest network and keep outpacing it, block after block, forever.

Why Does Bitcoin Need Miners?

Mining can look wasteful until you see what it buys. Miners supply three things no single company or server could provide as credibly:

• Security: Reversing a confirmed payment or spending the same coin twice would require controlling more than half the global hashrate, the so-called 51% attack. At September 2026 hardware and energy prices, assembling that much compute would cost billions, and a successful attack would destroy the value of the coins it targeted. Honest mining simply pays better.
• Issuance: Mining is the only mechanism that creates new bitcoin. The schedule is fixed in code, publicly auditable, and enforced by every node, which is why bitcoin's supply is predictable in a way no national currency's is. You can verify current issuance yourself using the Bitcoin.com block explorer.
• Decentralization: Anyone with hardware and electricity can take part from anywhere, with no permission required and no headquarters to pressure. When China banned mining outright in mid-2021, then the largest mining hub on earth, the network kept producing blocks without interruption while hardware relocated. The United States has since become the dominant hub, accounting for 75.4% of reported hashrate in Cambridge's survey data.

The Bitcoin Halving and Block Rewards

Miner revenue has two components: the block subsidy of newly minted coins, and transaction fees. The subsidy is cut in half every 210,000 blocks, roughly every four years, in an event called the halving. This is the mechanism that bends the supply curve toward the 21 million cap.

As of September 2026, with block height past 965,000, roughly 20.08 million coins have been mined. That is about 95.6% of the maximum supply, leaving under 1 million BTC to be issued across the next 114 years. The final fraction of a coin is expected around 2140.

The long-term design assumes transaction fees gradually replace the shrinking subsidy as miners' primary income, and that transition has barely begun. Over the 24 hours around September 7, 2026, fees made up just 0.43% of total miner rewards, and the weekly figure has been hovering between roughly 0.6% and 0.8% through the summer. Fee income is small and has been shrinking as a share of revenue, while the subsidy still does almost all the work. Whether fee demand grows enough to fund network security over coming decades is one of the genuinely unresolved questions in Bitcoin, and the fee share of miner revenue is the live indicator worth tracking rather than assuming an outcome.

Ways to Mine Bitcoin

There is more than one way to take part, and the differences come down to who owns the hardware and who carries the risk.

Solo mining is where newcomers most often misjudge the odds. A single modern ASIC running 270 TH/s represents about one three-millionth of a 934 EH/s network, and mining alone it would take roughly 66 years on average to find a single block. Pools solve this by letting thousands of machines search together and share every reward, converting a once-in-a-lifetime jackpot into small, steady daily payouts.

Cloud mining has long been the most scam-dense corner of this industry. Guaranteed daily returns, referral bonuses for recruiting others, and polished dashboards with no verifiable hashrate are the standard red flags. If the promised returns were real, the operator would mine for itself rather than sell the opportunity to strangers.

Is Bitcoin Mining Profitable in 2026?

Profitability comes down to two variables: what you pay for electricity and how efficient your hardware is. Large operators with cheap industrial power are still making money in 2026. Most home miners on retail electricity rates are losing it.

Start with the industry's core yardstick, hashprice: expected daily revenue per petahash per second of mining power. Hashprice sat near $39.63 per PH/s per day in early September 2026, a jump of roughly 22% over the preceding month from about $32.42, driven mainly by bitcoin's price recovery rather than any change in mining itself. Bitcoin traded around $78,350 on September 8, 2026, down significantly from its all-time high of $128,198 recorded on October 6, 2025.

The network data tells the story of a difficult year. Difficulty opened 2026 near 146.47 trillion and has since fallen to 127.45 trillion, while hashrate sits roughly flat near 934 EH/s as machines rotate in and out. Hardware has been switched off because it could not cover electricity costs, and efficient new machines have replaced rather than added to it. The 2024 halving cut gross revenue in half overnight, and the industry has spent the time since competing away what remained.

Two numbers decide which side of the line you land on:

• Electricity price, in dollars per kilowatt-hour (kWh). This is the dominant ongoing cost, frequently more than 60% of total operating spend. Cambridge's survey found energy price volatility is miners' single biggest reported concern.
• Hardware efficiency, in joules per terahash (J/TH). Treat it as fuel economy: how much energy a machine burns per unit of hashrate produced. Current-generation ASICs run around 13 to 15 J/TH. Machines from a few years back burn roughly double that for the same output.

Average US residential electricity price per kilowatt-hour. Home miners pay close to this rate, while large operations negotiate industrial contracts at a fraction of it. Source: US Bureau of Labor Statistics via FRED, St. Louis Fed.

Here is a worked example using a current-generation ASIC rated at 270 TH/s drawing about 3,650 watts (13.5 J/TH), at the September 2026 hashprice of $39.63 per PH/s per day:

Same machine, same network, opposite outcomes. The industrial column is also before hardware cost, cooling, maintenance, pool fees, and downtime, on a unit that typically costs several thousand dollars and becomes obsolete within three to five years. Note too that the same machine at July's hashprice of about $31 would have shown a much thinner industrial margin and a deeper residential loss. Revenue moves week to week while your electricity contract does not.

This is why the cost to mine 1 bitcoin varies so dramatically. For efficient large-scale operators with power under $0.05/kWh, all-in production cost sits comfortably below market price. For a household on retail rates, producing one BTC costs far more than simply buying one. Running the same 270 TH/s machine at September 2026 rates, gross revenue alone would take about 20 years to accumulate a single whole bitcoin.

Practical takeaways as of September 2026:

• Industrial miners with sub-15 J/TH fleets and power under roughly $0.08/kWh remain profitable, with margins that shift every difficulty epoch.
• Home mining at typical residential rates across most of North America and Europe loses money on electricity alone.
• Small home devices still make sense as education, supplemental heat, or a lottery ticket. They do not make sense as income.
• Run your own numbers before spending anything, using current difficulty and hashprice, and assume both will keep moving after you buy.

Nobody should treat mining as a guaranteed return. It is a capital-intensive business with volatile revenue, and well-funded public companies have gone bankrupt doing it.

How to Start Mining Bitcoin

If you have run the math and still want in, here is how to mine bitcoin step by step.

1. Do the profitability math first: Find your exact electricity rate on your utility bill, choose a specific machine, and model it against current difficulty and hashprice. If the daily margin is negative today, stop here or choose the hobbyist route.
2. Set up a bitcoin wallet: Rewards need a destination. A wallet where you control the private keys is strongly preferable to leaving payouts sitting on a third-party platform.
3. Choose hardware honestly: A competitive ASIC runs roughly $2,000 to $6,000, and needs serious ventilation plus often a dedicated electrical circuit. A small open-source device in the $100 to $300 range is quiet, cheap to run, and fine for learning. What does not work: phones, ordinary laptops, and gaming PCs. Bitcoin mining outgrew general-purpose processors more than a decade ago, and any mobile app claiming otherwise is simulating the process or harvesting something from you.
4. Join a mining pool: Unless you are running a warehouse, pool mining is the only route to regular payouts. Compare pools on fee structure, payout scheme, minimum payout threshold, and operating history.
5. Configure and monitor: Modern ASICs ship ready to run: plug in, open the machine's dashboard in a browser, enter your pool details and payout address. After that, mining becomes an operations job of watching temperatures, uptime, and power draw.
6. Keep records for tax: The U.S. Internal Revenue Service treats mined digital assets as income at their fair market value when received, with a separate capital gain or loss when you later sell. Most jurisdictions follow similar logic. Log every payout with its date and market price from day one, because reconstructing it later is painful.

How Much Energy Does Bitcoin Mining Use?

A great deal, by design. The debate is about whether that energy buys something worthwhile and where it comes from.

The most rigorous public measurement comes from the Cambridge Centre for Alternative Finance, which maintains the Cambridge Bitcoin Electricity Consumption Index as a live estimate. Its Digital Mining Industry Report, published in April 2025 and drawn from 49 firms representing roughly 48% of global hashrate, put Bitcoin's annual electricity use at approximately 138 TWh, around 0.5% of global consumption. In the United States, the Energy Information Administration estimated in early 2024 that cryptocurrency mining accounted for between 0.6% and 2.3% of national electricity consumption.

The same Cambridge survey found sustainable sources supplied 52.4% of mining's electricity, comprising 42.6% renewables and 9.8% nuclear, up from 37.6% in 2022. Natural gas at 38.2% has replaced coal, which fell from 36.6% to 8.9% over the same period. Miners are also unusual electricity customers, since machines can power down within seconds and some grid operators pay for exactly that flexibility during peak demand.

None of this settles the argument, and the figures move with hashrate, price, and geography.

Watch: Bitcoin Mining, Energy, and the Race for Power

At Consensus Miami in June 2026, David Sencil spoke with Eric Trump, co-founder and chief strategy officer of American Bitcoin, about where mining sits in the wider competition for cheap electricity. The 30 minute conversation covers mining's dependence on energy infrastructure, the concentration of hashrate in Texas, competition from data centers running artificial intelligence workloads, and how institutional capital has reshaped the industry since the ETF approvals. He speaks as an operator with a commercial stake in the sector, and the views expressed are his own.

Trade-Offs and What Comes Next for Bitcoin Mining

The defining story of mining in 2026 is consolidation under pressure. The 2024 halving compressed margins, the hashprice slump forced inefficient hardware offline, and public miners increasingly split their megawatts between bitcoin and artificial intelligence workloads. Mining companies are discovering their real asset was never the ASICs, it was the energized data center capacity, and competition for that power is now a structural feature of mining economics.

The trade-offs are real on both sides. Mining anchors the most secure open monetary network ever built, monetizes stranded and wasted energy, and supplies grid operators with flexible load. It also consumes serious electricity, concentrates in regions with cheap power, depends on an oligopolistic hardware supply chain, and pushes small participants out economically. Both lists are true simultaneously.

Three things worth watching from here: the fee share of miner revenue, which must eventually grow to fund long-term security; whether hashrate breaks decisively above the zettahash line after flirting with it in September 2026; and the 2028 halving, which cuts the subsidy to 1.5625 BTC and reruns this entire stress test at half the reward.

Conclusion

Bitcoin mining is the Proof of Work engine that confirms transactions, secures the blockchain, and mints every new coin on a schedule nobody can alter. Strip away the hardware catalogs and the energy arguments and the function is simple: mining converts real-world electricity into a ledger no single party can rewrite.

What has changed is who can do it economically. The 2024 halving cut the subsidy to 3.125 BTC, and the period since has pushed out operators without cheap power or efficient machines. As of September 2026, difficulty sits near 127.45 trillion, hashrate close to 934 EH/s, and revenue per unit of hashing power well below its 2025 peak. Mining at scale has become an energy and infrastructure business competing for the same megawatts as artificial intelligence, and the ASICs are the smaller half of the investment.

Unless your electricity costs meaningfully less than retail, buying bitcoin outright is cheaper than producing it. Competition has pushed production costs down to the point where operators with the cheapest power set the floor, and that same pressure keeps hardware efficiency improving. Home mining still earns its place as a way to understand how Bitcoin works, contribute hashrate to it, and in some setups reuse the heat.

Miners have absorbed four halvings already and kept producing blocks through each one. The harder long-term question is fee revenue. Transaction fees still account for under 1% of what miners earn, and the protocol's design eventually needs them to account for nearly all of it.