New editorial guide, published 16 October 2026. The historical sources linked below are preserved separately and are not represented as current product or service information.
A Bitcoin wallet displays a convenient total, but the network does not store that total as a bank-style account balance. It records outputs created by transactions and whether those outputs have subsequently been spent. Understanding this difference makes change addresses, transaction fees and apparently complicated explorer pages much easier to read. This new guide revisits the subject of a recovered BitCryptoNova article without carrying forward its dated shopping examples or treating an analogy as a literal description of the protocol. The useful question is not where a digital coin sits, but which outputs a future transaction can spend.
An output is an instruction plus an amount
A transaction output specifies an amount and spending conditions. A later transaction can reference that output and provide the information needed to satisfy those conditions. Until it is spent, it belongs to the unspent transaction output set, usually abbreviated UTXO. The familiar idea of a coin is useful for conversation, but the underlying object is more precise: a particular output from a particular transaction, identified by the transaction identifier and its position among that transaction's outputs. Several outputs can represent funds controlled through the same wallet.
This distinction explains why an address is not a little container that physically holds moving coins. An address encodes information used when constructing an output. A wallet discovers relevant transactions, tracks which outputs remain available and works out which spending conditions it can satisfy. Its displayed balance is the result of that accounting. Different wallet views can legitimately distinguish confirmed, unconfirmed, watch-only or otherwise restricted funds. Before interpreting a total, establish what that particular screen includes and whether every included output is presently spendable.
Spend the whole input, create fresh outputs
Consider a deliberately simplified example. A wallet can spend one output worth 100,000 satoshis and wants to pay someone 30,000 satoshis. It constructs a transaction consuming the original output and creating a payment output, a change output and an implied fee. With a 1,000-satoshi fee, the change would be 69,000 satoshis. These numbers illustrate the accounting only; they are not a recommended fee. The original output does not remain partly unspent. New outputs replace the amount allocated from it.
The recipient's new output is also unspent until a later transaction consumes it. Calling it a spent output merely because it was the payment leg would confuse the transaction's inputs with its outputs. The payment consumes an older output; the payment itself creates a new one. A useful drawing therefore has arrows from old outputs into the transaction and arrows from the transaction into new outputs. Label the fee separately as the difference between total input value and total output value, rather than inventing a fee output.
Why change can use a new address
Wallets commonly generate a different address for change. Seeing an unfamiliar address in your own transaction is therefore not automatically evidence that funds went to an unknown person. The wallet may control the corresponding spending information. Conversely, an explorer cannot simply declare every second output to be change with certainty. It may apply a heuristic based on script types, address history or transaction structure. Those patterns can be informative, but they are not the same thing as cryptographic proof of ownership.
This is a reason to verify a payment through the wallet's own transaction details as well as a public explorer. Identify the amount, intended recipient and change classification, and understand whether the wallet is displaying a net change in your balance or the transaction's gross output value. A transaction with several outputs can look much larger than the payment you remember making. The difference is usually an accounting question to investigate, not a reason to assume that the explorer or wallet must be concealing something.
Coin selection changes the shape of a payment
A wallet may control many small outputs rather than one large output. To fund a payment it chooses a subset, a process called coin selection. More inputs can increase transaction size and consequently the fee required at a chosen fee rate. This is why the amount being sent does not, by itself, determine a Bitcoin transaction's network fee. Two payments for the same amount can require different amounts of transaction data. Script type and the exact construction also matter, so input count is an explanation rather than a complete calculator.
Selecting outputs also affects what observers can infer. When a transaction spends several inputs together, analysts may infer a relationship among them. Collaborative transaction arrangements can undermine that inference, so it should not be treated as an ownership certificate. For the wallet user, the practical lesson is that combining outputs can change both cost and the public information revealed by a payment. A good interface explains these trade-offs instead of presenting coin selection as a magic setting that always produces one objectively best transaction.
Small outputs are not automatically useful funds
An output can exist while being inconvenient to spend economically. If spending it requires a fee comparable to its value, including it in a transaction may make little sense under current conditions. The word dust is often used loosely in tutorials, but protocol validity, relay policy and personal economic usefulness are different questions. A tiny number appearing in an explorer should not be confused with a guaranteed usable payment. Look at the wallet's estimate and the relevant spending conditions before drawing conclusions about its practical value.
Similarly, an unsolicited small transfer does not create an obligation to interact with a sender or follow a message attached elsewhere. Blockchain records can be examined without opening promotional websites or importing unknown wallet data. Keep the observation separate from the action someone hopes to provoke. A useful personal record notes the transaction identifier, date observed, amount and reason for considering the output relevant. This preserves enough information for later review without turning every unfamiliar transaction into an instruction that must immediately be acted upon.
Read the accounting before the story
When examining a transaction, begin with inputs, outputs and their values. Check whether you are looking at the same network and the exact transaction identifier, then identify confirmation status and the wallet's interpretation. Only after that should you ask who might control the addresses or what broader economic activity occurred. A transfer can represent change, internal accounting, a batch payment or a more elaborate construction. Public transaction structure establishes some facts directly while leaving many human explanations unresolved. Separating those layers is the core skill.
The older BitCryptoNova source is preserved alongside this guide so readers can compare the explanation and the historical wording. Its references to particular merchants and dated purchasing arrangements are not evidence that those arrangements exist today. The enduring idea is simpler and stronger: an input consumes an output, a transaction creates new outputs, and the wallet builds a useful balance from the outputs it tracks. Once that model is clear, change stops looking like money disappearing and starts looking like ordinary transaction accounting.
Technical references
From the recovered archive
- UTXO: The Complete Guide to Bitcoin “Change” for Beginners 2025 · captured 9 October 2025