How Does Blockchain Technology Work? A Plain Answer

A bank keeps a ledger, which is just a list of who owns what and who moved money to whom. The bank owns that list, keeps it on its own servers, and everyone has to trust it. The bank says a payment happened, and you accept it because the bank runs the records.

A blockchain does the same job with a different owner: everyone. Instead of one company holding the list, thousands of independent computers each hold a full copy, and they agree on every update using rules instead of trust (wikipedia.org). The idea grew directly out of the 2008 financial crisis, when builders watched one company's failure freeze everyone's records. Their answer was a record no single failure could take down. You can verify the logic yourself with nothing but pen and paper: write a list of ten purchases, compute a simple checksum for each line and for the whole page, and watch how one erased entry breaks every check when you try to reconcile it.

The key insight is separation of duties. A bank is one company, and when it makes a mistake, goes offline, or gets hacked, everyone notices the same failure at once. A blockchain spreads the records so widely that no single crash can erase them. Nobody presses a reset button because nobody owns the button. That redundancy is expensive and slow, which is why blockchains only make sense for records where trust matters more than speed.

Transactions get bundled into batches called blocks. Think of one block as one page in a ledger. Bitcoin has added a new page about every ten minutes since January 3, 2009, when the first block, called the genesis block, was created (wild.io). Other chains move much faster, with Solana producing a new block about every 400 milliseconds and Ethereum landing one roughly every twelve seconds. These pages pile up in order, which is where the name comes from: a chain of blocks.

1. A wallet user signs a transaction with their private key, a secret password that proves ownership without revealing itself
2. The signed transaction is broadcast to the network, hopping between computers in seconds
3. It waits in the mempool, the shared waiting room of unconfirmed transactions
4. Builders pick transactions from the waiting room, usually highest fee first, and pack them into a block
5. The block is added, giving the transaction its first confirmation
6. Each new block built on top adds another confirmation, which is how merchants know the payment is sticking

Each block holds two kinds of data. The body carries the actual transactions, summarized into a fingerprint called the Merkle root, invented by Ralph Merkle in 1979 (blockchainpeople.nl). The header carries bookkeeping details: a timestamp, a number called a nonce, and the most important part, the hash of the block before it. The header is small, but it contains everything needed to verify the chain's integrity without reading every transaction inside every block.

A hash is a cryptographic fingerprint. You feed any input, a single word or a million transactions, into a hash function and it crunches the input into a short, fixed-length string. Bitcoin uses one called SHA-256 (wild.io). Two properties make hashes magical: the same input always produces the same fingerprint, and changing even one character of the input changes the fingerprint completely, with no resemblance to the old one.

Because every block's header includes the hash of the previous block, the pages are mathematically locked in order. Follow the dominoes. Someone edits a transaction inside block 100. Block 100's hash instantly changes. But block 101 recorded the old hash of block 100, so block 101 no longer matches either. Fixing block 101 changes its hash, which breaks block 102, and so on all the way to the newest page (wild.io). One small edit forces a rewrite of everything after it.

This is also why hashing doubles as a certificate of integrity. When a Bitcoin wallet tells you a block has fifty confirmations, it means that block has been buried under fifty subsequent blocks, and every single one carries a hash derived from the block you want to undo. Each confirmation is another layer of sealed math on top of your transaction.

A network full of strangers still needs to agree on which block comes next. The rule set that does this is called consensus, and there are two big families (ethereum.org). Proof of work, used by Bitcoin, has computers compete to solve a costly math puzzle, and the winner proposes the next block. Proof of stake, used by Ethereum since the Merge upgrade in September 2022, has participants lock up their own coins as collateral for the right to propose blocks, and cheating destroys their stake, a penalty called slashing (blockchainpeople.nl).

The effort is the point. On both designs, telling the truth is the only move that pays and cheating is a money-loser. Ethereum's switch slashed its energy use by an estimated 99.95% compared to its old proof of work design, according to the Ethereum Foundation (blockchainpeople.nl). Solana takes a different approach again, using proof of history as a timestamping layer to speed up consensus, showing that the mechanism is a design choice with tradeoffs, not a one-size-fits-all answer.

Neither model is perfect. Proof of work critics point to the massive energy bills. Proof of stake critics argue that those who hold the most coins have the most influence, which looks a lot like the centralized system blockchain was supposed to replace. Both critiques are partly true, and the tradeoffs are why no single blockchain dominates every use case.

You can edit your own copy of any blockchain, sure. But thousands of other computers still hold the original, unbroken chain, and the network follows whichever copy a majority agrees on. To make an edit stick you would have to redo the math on every block after the tampered one, then keep pace with the whole network while it keeps adding new honest blocks.

Work the numbers and you see why nobody bothers. In mid-2026 Bitcoin's network was running at roughly 830 exahashes per second, which is hundreds of quintillions of calculations per second (wild.io). Consistency is why each block built on top of a transaction adds another confirmation, and why large Bitcoin payments are traditionally treated as solid after about six confirmations. Reversing something six blocks deep means redoing six blocks while outpacing the entire world's computing power (wild.io).

Even a well funded attacker would need hardware, electricity, and time that costs more than any likely payoff. In practice, the largest known attacks on blockchain networks have targeted the apps and exchanges sitting on top of the chain, not the chain itself. The ledger held. The exchanges that failed are the ones we read about in the news.

Crypto is the obvious use, but the ledger is generic. Banks test blockchains to settle international payments faster. Supply chains use them to prove where a coffee bean or a diamond came from. Land registries use them to make property records hard to fake. In every case the pitch is the same: a record that no single party owns and nobody can quietly rewrite (ethereum.org).

The cost is speed and redundancy. Every computer holds a full copy, so storing a gigabyte of data on a blockchain means every node stores it, which is slow and expensive. That is why blockchains are used for small, valuable records like balances and contracts, not for photos or files.

• Supply chain tracking: proving origin, handling, and transit of goods
• Digital identity: giving people control over their own credentials
• Tokenized assets: representing real estate, art, or securities as digital tokens
• Voting systems: creating transparent, auditable records of who voted
• Smart contracts: self-executing agreements that run when conditions are met

The private blockchain debate is worth touching on, because most enterprise uses are not fully public. A consortium chain like those used by shipping networks lets a group of known companies share one ledger with permissioned access, getting the tamper-evidence at a fraction of the public chain's redundancy cost. Purists argue this sacrifices the decentralization that makes the technology interesting. Practitioners point out that a shipping giant does not want its invoices public to the whole world. Both views are reasonable, and the split explains why the public blockchains get the headlines while the private ones quietly run inside supply chains and clearinghouses.