What is Bitcoin? An explanation from scratch

For thirty years, the internet made almost everything cheaper—email, calls, books—but it could never remove the bank from the middle when you moved money. Before talking about price, investment, or mining, it is worth understanding the specific problem Bitcoin solved in 2009 by breaking that rule. This article builds it step by step, starting from the trust that sustains any exchange of value and reaching what Bitcoin does—and does not do—as a system, without taking anything for granted. It is the first in a series of three: here is the "why"; the following ones build what a generic blockchain is and what DeFi is.

Why does paying by card require trusting others?

When you pay by card in a coffee shop, something happens that is barely noticeable. The coffee shop does not check if you have money in the bank—nor can it, as it has no access to your account. What it does is ask a third party: your card network (Visa, Mastercard), which in turn asks your bank. If everyone says "yes," the transaction goes through. If at any point in that chain someone says "no," there is no coffee.

Neither you nor the coffee shop truly trust each other. What sustains the payment is a network of intermediaries who trust each other. You pay with that borrowed trust—and you pay literally, because each intermediary charges a fee for participating.

That architecture has four costs that we rarely see:

1. Friction — commissions, delays, currency conversions, banking hours.
2. Control — Visa, your bank, or a government can block a payment they don't like, even if it is legal.
3. Single point of failure — if the infrastructure provider on which payments depend stops responding, your money is frozen even if it exists. The causes are varied and real: a faulty update (CrowdStrike, July 2024, took down banks, airlines, and hospitals in hours), a cyberattack (the ransomware against ICBC in November 2023 paralyzed part of the US Treasury bond market settlement), or a physical attack against data centers in a conflict zone, as seen in the war between Iran and Israel. Your account doesn't need to be compromised nor does your bank need to go bankrupt: it is enough for one link in the chain to fail.
4. Exclusion — around 1.3 billion adults in the world do not have a bank account; for them, that network simply does not exist.

For thirty years, the internet made almost everything cheaper—email, calls, books, movies—but it could never eliminate the money intermediary. The reason is technical: if I send you a PDF, I can send it to another person at the same time without anyone noticing. If I send you 10 digital euros and no one is keeping track, I can "spend" those same 10 euros with someone else at the same time. This is called the double-spending problem, and for decades no one knew how to solve it without someone keeping a central tally. That is why, until 2009, digital payments always had a bank behind them.

What changed in 2009 was Bitcoin. And before talking about Bitcoin, you have to understand the trick it invented: a way of keeping track without anyone in particular keeping it.

What changes if no one controls the application?

Today we all use banking applications. Behind any of them is a program running on the bank's servers and a database that stores who has how much. When you make a transfer, the application receives the order, the program processes it, and the database is updated: your balance goes down, the recipient's goes up.

That scheme has a feature that no one questions because we have been using it for decades: a single entity controls the program and the database. The bank decides what is executed, when, and for whom. If it wants to block an account, it blocks it. If its server goes down, payments stop. If it decides to change the rules, it changes them.

A blockchain is exactly the same thing—an application with its database—but with two important differences:

1. No one in particular controls the program or the database. The same software is running simultaneously on thousands of independent computers spread across the world. Anyone can join and set up their own: these participants are called nodos. They constantly synchronize with each other to maintain a single consistent version of the database. If a node tries to record a lie in its copy, the others do not accept it: its version is sidelined and stops counting.

This is not a metaphor: in April 2026, the Bitnodes tracker counted around 23,000 reachable nodes (those that accept public connections; the real total is higher, as many run behind firewalls). The country with the most was the United States, with about 2,700—less than 12% of the total, so no single territory concentrates a majority. A significant portion of these nodes operates over the Tor network so that not even their internet provider can locate them. Each one keeps a full copy of the database, which by mid-2026 weighs around 750 GB and grows by a few GB per month. That is the physical cost of having no owner: massive redundancy instead of a central server.

2. Anyone can use that application to create their own point system. Just as on a social network anyone can open an account and post whatever they want, on a blockchain anyone can create a new digital asset—let's call it a "point," because that's what it does in the end—and let it circulate within that same database. The first point system that lived on the first blockchain is called bitcoin (lowercase: the asset). The blockchain itself is called Bitcoin (uppercase). The design allows for many more point systems on top, an idea that was only seriously deployed years later when a second blockchain called Ethereum appeared (this is the subject of the next article in the series).

These two features—ownerless application and many point systems on top—together solve the double-spending problem: no user can spend the same thing twice because all nodes are seeing the same balance at once, and each "point" has its own clear accounting within the application. But they open two obvious questions:

• Why would anyone go to the trouble of maintaining one of those thousands of nodes? Bitcoin solves this by paying whoever does it, in bitcoins. This task is called mining, and those who do it are mineros.
• What if someone with money sets up thousands of fake nodes to impose their version of the database? Bitcoin solves this by making maintaining a node cost real electricity—so much that faking the database is more expensive than the money that could be stolen by manipulating it. This artificial cost is called proof of work.

The combination of both mechanisms is what sustains the system: an application that works because whoever wants to participates, and cheating costs more than playing fair. And within that application, many different point systems can live, each with its own rules.

Why is Bitcoin not an application but an architecture?

We have talked about the blockchain as if it were an application because it is the closest intuition. Technically it is not just that: it is an architecture—a set of common rules for moving and recording value that many participants follow at the same time.

The closest reference you probably already have is SWIFT: the network through which the world's banks send the messages needed when you order an international transfer. SWIFT is not a bank, it has no accounts, it does not lend money. It is the common protocol that allows thousands of heterogeneous banks to understand each other—each runs its own software and serves its customers, but all respect the same format. Common protocol, many participants, no single operator: that is what a blockchain shares with SWIFT.

The decisive difference is that SWIFT is not neutral. There is a committee that decides which banks enter, which meet the requirements, and which are left out; in 2022, the main Russian banks were expelled as a geopolitical sanction and stopped processing international transfers overnight. SWIFT works because its participants trust each other and have been previously approved. It is architecture, but architecture with a gatekeeper.

A blockchain has no gatekeeper. There is no committee to decide who sets up a node or who can receive a payment: anyone with an internet connection and the software can participate, and the protocol processes all transactions under the same rules. This property is called neutrality or censorship resistance. In practice, it is what allows activists in countries with banking blockades to receive donations, or a migrant to send remittances bypassing the average 6-8% charged by conventional channels.

The other classic analogy—the gold standard—captures that angle better. Physical gold did not discriminate either: whoever had gold paid with it without asking anyone's permission. Its limitation was operational: it is heavy, requires vaults, travels poorly across borders, and can be confiscated. And moving the piece is moving the ownership—the gold you have is yours in a literal sense, not a bank's promise to deliver it to you later. A blockchain digitalizes those two properties—neutrality and ownership without an intermediary—without the physical restrictions.

This combination—neutral architecture where the value is the asset, not a promise—clarifies why there are so many blockchains. Bitcoin, Ethereum, Solana, and the rest are not competing applications: they are different architectures with their own rules (type of nodes, processing speed, cost per transaction, what can be built on top). They compete as SWIFT, the gold standard, and a new payment network would compete: different rails to move value. Hence, points created on one blockchain do not automatically appear on another; they live in different architectures and to move them, bridges must be invented.

When has it mattered in practice that Bitcoin has no gatekeeper?

The property of operational neutrality sounds abstract until you look at where it has been tested. Four documented cases, in chronological order, help to see where utility stops being theoretical:

1. Wikileaks, December 2010. After leaking diplomatic cables from the US State Department, Visa, Mastercard, PayPal, Western Union, and Bank of America suspended donations to the organization without prior judicial indictment, due to political pressure. Bitcoin donations remained operational and became one of the organization's main financial channels for years. It is the first documented case where the "non-censurable" property was tested on an institutional scale.
2. Canada, February 2022. The government invoked the Emergencies Act and froze more than 200 bank accounts linked to the so-called Freedom Convoy without a court order. GoFundMe reversed the donations collected on its platform. Bitcoin donations (approximately $1 million) could not be stopped at the protocol level; the only way to neutralize them was to identify the recipients and go after them individually. It is the first case where a G7 country explicitly used banking infrastructure as a sanctioning tool against its own citizens.
3. Russia, March-April 2022. Following the mass exit of Visa, Mastercard, PayPal, major Western banks, and the partial disconnection from SWIFT, Russian citizens not individually sanctioned were practically cut off from international financial infrastructure. According to data from Chainalysis and P2P operators, ruble-crypto volume multiplied between three and five times in a few weeks. It is the case of a full-state geopolitical sanction scale.
4. Nigeria, 2023-2024. The Central Bank of Nigeria limited the purchase of hard currency to symbolic amounts (around $20/month) and promoted the eNaira, a central bank digital currency that failed to gain traction. In parallel, approximately 33 million adults—about 15% of the adult population—consolidated the use of USDT and BTC via P2P as a de facto parallel market. It is the case of mass adoption scale under capital controls.

The utility of Bitcoin is better understood by reading these four cases than with abstract arguments. The progression is clear: from an alternative channel for an organization (2010), to a vehicle for social movements in mature democracies (2022), to an individual refuge from geopolitical sanctions (2022), to parallel financial infrastructure for tens of millions of people (2024). None of the four situations would have been resolved by a traditional banking system—by definition. And none requires speculating on the price of the asset: operational utility precedes any valuation discussion.

What does Bitcoin solve and what does it not solve?

With the mechanism clear, it is worth separating what Bitcoin does from what is attributed to it without it actually doing it. Compared to the four costs of traditional banking we opened with, the breakdown is as follows:

The table reveals the pattern: every advantage of Bitcoin comes with its exact counterpart. The irreversibility that prevents you from being censored is the same one that prevents undoing a mistake. The honest list is shorter and more useful than the mysticism that usually accompanies it.

What it does solve:

• Double-spending without an intermediary. It is the foundational problem. The shared database and proof of work solve it mathematically, without the need for a bank to keep the central tally.
• Operational neutrality. No one can block, reverse, or censor a valid transaction. This property matters structurally for those living under capital controls or sanctions, and little for those with stable banking.
• Ownership without an intermediary. If you hold the keys to a bitcoin, that bitcoin is yours in a literal sense, not a custodian's promise to return it to you. It is the digital version of gold's "move the piece, move the ownership."
• 24/7 final settlement. A confirmed transfer is irreversible and operates every day of the year at the same time, without holidays, cut-off windows, or interbank closures.
• Fixed and verifiable monetary policy. The supply is programmed with a cap of 21 million bitcoins. As of mid-2026, approximately 19.8 million have been mined. Chainalysis and other on-chain analysts estimate that between 3 and 4 million (approximately 17-20% of the supply) are in addresses that have not moved for more than seven years and are, for the most part, unrecoverable—lost keys, deceased without digital inheritance, wrong transfers to ownerless addresses. The effectively liquid supply is therefore significantly lower than the total mined. Issuance is halved approximately every four years (the so-called halving): after the April 2024 halving, annual issuance is around 164,000 new BTC. No more is "printed" by political decision; the code prevents it and any node can verify it independently.

What it does not solve—or only partially solves:

• Privacy by default. Bitcoin is pseudonymous, not anonymous. All transactions are public and traceable on the blockchain; the only thing it hides is the real identity behind each address, until someone links it to a name. Decent forensic analysis reconstructs much of the trail.
• Speed and scale on-chain. The base layer processes between 7 and 12 transactions per second, with confirmation times of minutes. For coffee payments, that is not enough; that is why Lightning exists, a second layer on top of Bitcoin that bundles many payments off-chain and settles only the result on it, achieving instant payments at near-zero cost.
• Price stability. Bitcoin is volatile. That is neither good nor bad in the abstract; it is real. Treating it as a checking account for daily expenses is ignoring that volatility. Treating it as a long-term position makes it matter less.
• Reversibility of errors. If you send bitcoins to the wrong address, there is no one to call. If you lose the keys, there is no way to recover them. Freedom and responsibility are the same thing.
• Customer service. There is no 1-800 number for Bitcoin. The protocol has no office; the services built on top (exchanges, wallets, custodians) do, but those services are third parties with their own risks, not part of Bitcoin.
• Global inflation, poverty, or inequality. Bitcoin is a tool, not a social movement. It solves specific problems for those who have them; it does not transform economies on its own.

The "what it doesn't" list is longer than the "what it does" list. That is not a flaw: it is honesty about the invention's perimeter. Bitcoin solves a specific technical problem—double-spending without an intermediary—and from that solved problem derive a few very useful properties for very specific scenarios. Everything else is applications built on top, or expectations that the tool cannot satisfy.

When does Bitcoin make sense for a normal person?

Three honest questions. If any of the three is answered "no," the reasonable conclusion is that Bitcoin probably doesn't add much to your life today—and that's perfectly fine.

1. Is your banking and jurisdictional situation stable? If you live in a country with solvent banking, freedom of capital movement, respected property rights, and you don't anticipate a change in any of the three, the operational utility of Bitcoin for you is marginal. Your checking account meets your needs. That doesn't mean Bitcoin has no value; it means the value it has matters more in inverse proportion to the quality of the financial infrastructure you already enjoy.

2. Do you want exposure to the store of value category? Bitcoin belongs in the same drawer as gold, investment art, and certain commodities: assets that do not generate cash flow and are only worth what someone is willing to pay. That category is legitimate, but not mandatory. Deciding if you want exposure to it is a portfolio decision that precedes any discussion about the specific price of Bitcoin. If you are interested in delving into how that decision is reasoned, it is covered in Should I buy Bitcoin? The honest answer.

3. Can you take on the responsibility of custody? Holding bitcoins in self-custody means that you—and no one else—are responsible for protecting the keys. If you lose them, the bitcoins are lost. If they are stolen, the bitcoins are gone. There are alternatives (professional custodians, ETFs, regulated exchanges) that delegate that responsibility and, with it, part of the very premise of the invention. Both options are reasonable for different profiles; pretending otherwise is selling smoke.

If after the three answers the "yes" still stands, the next steps are below. If it has fallen through on any, the rational decision is not to enter, and it is worth also reading Why crypto may not be for you before continuing.

Where do I go from here?

This is the first piece in a series of three. The next article—Crypto from scratch—builds on what we have established here and addresses what a generic blockchain is, why there are so many, what Ethereum, tokens, stablecoins, and NFTs are, and why points created in one architecture do not move natively to another. The third—DeFi from scratch—closes the series by explaining how, on those rails, complete financial services are built without intermediaries.

If you want to go deeper before continuing with the series, complementary readings:

• Why crypto exists — the argumentative "why," with specific cases of banking crises (Argentina 2001, Cyprus 2013, Lebanon 2019, Venezuela) that motivated much of the real adoption.
• Should I buy Bitcoin? The honest answer — direct pros and cons for the personal decision to buy or not.
• Crypto vs. traditional banking — the point-by-point comparison between the two rails for moving money.
• Why crypto may not be for you — the personal check-up before any operational step.

Sources and external references: Bitcoin whitepaper (Satoshi Nakamoto, 2008) · World Bank — Global Findex · World Bank — Remittance Prices Worldwide · Chainalysis — Global Crypto Adoption Index · Bitnodes — reachable Bitcoin nodes · Wikileaks