Why Bitcoin On-Chain Transfer Value Can Vary by Up to 6x, BIS Study Finds

Bitcoin’s public blockchain records every confirmed transaction, making it one of the most transparent financial networks in the world. But turning that raw blockchain activity into an accurate measure of economic value is more complicated than simply adding up transaction outputs. A September 2026 study from the Bank for International Settlements found that estimates of Bitcoin on-chain transfer value can vary by as much as sixfold, depending on the methodology used to interpret transaction data. The distinction between recorded transfers and on-chain volume is therefore important when assessing how much economic activity is actually taking place.

The finding highlights an important distinction between technical blockchain activity and actual economic transfers between different participants. Bitcoin’s UTXO structure, change outputs, self-transfers and wallet-management activity can all influence reported transaction values. The BIS research does not suggest that Bitcoin transactions themselves are inaccurate. Instead, it shows that analysts can reach very different conclusions from the same blockchain when they use different assumptions about what should count as genuine economic activity.

Understanding this difference matters for investors, researchers and anyone using Bitcoin on-chain data to assess network activity. It also helps explain why transfer-volume estimates from different analytics platforms do not always match, even though they are analyzing the same underlying blockchain.

Bitcoin’s on-chain transfer value can look very different depending on how blockchain transactions are measured. The BIS study found that estimates of Bitcoin transfer activity can vary by as much as sixfold across different measurement methods. The difference does not mean Bitcoin transactions are inaccurate. Instead, it reflects the difficulty of separating genuine economic transfers from technical movements of BTC recorded on the blockchain.

The main reason is Bitcoin’s UTXO, or Unspent Transaction Output, model. When someone spends Bitcoin, the entire input is consumed and new outputs are created. If the input is larger than the amount being sent, the remaining BTC is typically returned to the sender through a change output. For example, a user spending a 1 BTC input to send 0.2 BTC may receive roughly 0.8 BTC back as change. A simple calculation that counts every output could record close to 1 BTC of transfer activity even though only 0.2 BTC was economically transferred to another party.

Why Measurement Methods Produce Different Bitcoin Transfer Estimates

Different analytics methods therefore produce different estimates of Bitcoin transaction volume. Broader approaches may include most transaction outputs, while more conservative methods attempt to remove identifiable change outputs, self-transfers and other movements that may not represent new economic activity. Because change is often sent to newly generated Bitcoin addresses, identifying which outputs belong to the original sender is not always straightforward.

This is why the BIS researchers caution against treating raw blockchain figures as perfect measures of economic activity. The study does not conclude that all Bitcoin transfer statistics are overstated by six times. Instead, it shows that the methodology used to calculate Bitcoin on-chain transfer value can significantly influence the final number, making transparency about measurement methods essential when comparing on-chain data.

Bitcoin transactions are built from unspent transaction outputs, or UTXOs, rather than conventional account balances. Each Bitcoin UTXO represents a specific amount of BTC that can be used as an input in a later transaction. Once spent, that output is consumed and replaced by one or more new outputs. This structure makes Bitcoin transparent and easy to verify on-chain, but it also makes raw transfer-value calculations more complicated than simply tracking money moving from one account to another.

The challenge is that blockchain data records every output created by a transaction, even when some of that value is simply returning to the original owner. Without additional analysis, those internal movements can make Bitcoin on-chain transfer estimates appear larger than the amount of BTC that actually changed economic ownership. This becomes particularly important when raw blockchain data is used to estimate payment activity, network growth or capital movement.

How Change Outputs Affect Bitcoin Transaction Volume

Suppose a wallet has a 1 BTC UTXO but wants to send only 0.25 BTC. The transaction may create one output of 0.25 BTC for the recipient and another output of roughly 0.75 BTC returning to the sender, minus transaction fees. From the blockchain’s perspective, both outputs are legitimate parts of the same transaction and remain permanently recorded.

If an analytics model counts both outputs equally, the transaction could appear to represent close to 1 BTC of transfer value. In economic terms, however, only 0.25 BTC was sent to another party, while the remaining BTC stayed under the sender’s control. Repeating this pattern across large numbers of transactions can create a sizable gap between raw output totals and estimates of actual economic activity.

This is why raw blockchain output totals should not automatically be interpreted as direct measures of Bitcoin payment activity. They show how much value technically moved between outputs, but not necessarily how much value changed hands between independent participants.

Why Change Outputs Are Difficult to Identify

Identifying change is not always straightforward because Bitcoin wallets commonly send change to a new address rather than returning it to the original sending address. This makes it more difficult for outside observers to determine which output belongs to the sender and which represents the actual payment.

Blockchain analytics providers therefore use heuristics and clustering techniques to estimate which outputs are likely to be changed. These methods are not perfect, and different providers can classify the same transaction differently. Some models remove more suspected change, while others use more cautious filtering to reduce the risk of excluding legitimate payments.

These methodological differences help explain why estimates of Bitcoin transaction volume can vary between analytics platforms. The blockchain itself remains consistent, but interpretations of its transaction structure can produce different economic estimates.

Self-Transfers and Wallet Restructuring Add More Noise

Change outputs are not the only reason Bitcoin transfer figures can appear unusually large. Users, exchanges, miners and custodians may move BTC between addresses they already control for wallet management, security, consolidation or operational purposes. Large institutions can therefore generate substantial on-chain activity even when economic ownership of the coins has not changed.

These self-transfers are genuine blockchain transactions, but they do not necessarily represent new activity between separate market participants. An exchange transferring Bitcoin between hot and cold wallets, for example, may create a large on-chain transaction without any associated customer purchase or sale.

For investors and analysts, the key point is that Bitcoin transfer volume depends heavily on how UTXOs, change outputs and self-transfers are classified. Raw blockchain records remain transparent, but interpreting them accurately requires a clear methodology rather than simply adding together every recorded output.

The BIS study shows that Bitcoin transfer value changes significantly depending on how transaction outputs are classified. Rather than relying on a single raw figure, the researchers compared progressively more selective approaches designed to separate broader blockchain movements from transfers more likely to represent economic activity.

1. Gross Output Value

The broadest approach counts the value of Bitcoin transaction outputs with relatively limited filtering. Because a single transaction can contain both a payment to another participant and BTC returned to the sender as change, this method can produce the highest estimate of on-chain transfer value. It captures a large share of activity visible at the output level, but it can also include value that remains under the same owner’s control.

2. Change-Adjusted Transfers

The second approach reduces the total by identifying outputs that are likely to represent change returned to the sender. This produces a more conservative estimate of Bitcoin transfer activity because it attempts to remove BTC that moved technically on-chain but did not necessarily move between separate economic owners. The difficulty is that change frequently goes to new addresses, requiring analysts to use additional assumptions to identify it.

3. Heuristic-Filtered Transfers

The most restrictive approach applies additional heuristics to identify probable change, self-transfers and other non-economic movements. This produces a lower estimate intended to get closer to transfers between distinct economic participants. The difference between broader and more heavily filtered calculations reached as much as sixfold, demonstrating how strongly Bitcoin on-chain transfer estimates can depend on the methodology used.

Bitcoin’s traditional market capitalization is calculated by multiplying the current BTC price by the circulating supply, meaning every coin is effectively valued at today’s market price. Realized capitalization uses a different approach by valuing each UTXO at the Bitcoin price when it last moved on-chain. Because many bitcoins were acquired or last transferred at much lower historical prices, Bitcoin realized cap can be significantly below conventional market cap. Coverage of the BIS research noted that Bitcoin’s standard market capitalization has at times been roughly four times higher than realized capitalization, illustrating how different valuation methods can produce very different perspectives on the network.

However, the gap does not mean Bitcoin’s traditional market cap is necessarily wrong or inflated by four times. The two metrics measure different things: market cap reflects the value of the circulating supply at the current BTC price, while realized cap is commonly used as an estimate of the network’s aggregate on-chain cost basis. Current Bitcoin market data can provide the market-price side of that comparison, while realized capitalization reflects historical prices at which coins last moved. Neither metric should automatically be treated as Bitcoin’s single “true” valuation.

The BIS study highlights an important limitation of Bitcoin on-chain data: blockchain records are transparent, but the economic meaning behind those records still depends on how analysts interpret them. Metrics such as transfer value, transaction volume and network activity can be useful, but they should not be treated as perfectly objective measures without understanding the methodology behind them. Two data providers can analyze the same Bitcoin blockchain and still produce noticeably different results because they apply different filters, assumptions and classifications.

For investors and researchers, this means methodology matters as much as the headline number. When comparing Bitcoin on-chain metrics across dashboards, reports or analytics platforms, it is important to check whether figures exclude internal wallet movements, apply address-clustering techniques or make other adjustments. A sudden increase in reported transfer value may reflect genuine economic activity, but it can also be influenced by custodial movements, wallet reorganizations or differences in transaction classification.

The broader takeaway is not that Bitcoin blockchain data is unreliable. Instead, the BIS research shows that raw on-chain activity should be interpreted with context and, where possible, compared with additional indicators such as active addresses, realized capitalization, exchange flows and transaction counts. Combining several metrics can provide a more balanced view of Bitcoin network activity than relying on a single transfer-value figure.

• One account for everything. Older platforms split your money across separate "spot," "margin," and "futures" accounts and expected you to understand why. KuCoin 5.0's unified account removes that entirely — deposit once, and everything is simply there.
• Stocks, indices, and commodities. KuCoin 5.0 expands beyond crypto into global markets. When crypto chops sideways and equities rally (or the reverse), you rotate in minutes instead of opening a brokerage account and waiting days for fiat rails.
• Real-world assets (RWA). Tokenized exposure to traditional assets like commodities, right inside your crypto account. One of the fastest-growing segments in global finance is no longer reserved for institutions — you access it from the same balance you trade with.
• Earn while you learn. Not ready to trade? KCUSD lets your stablecoins earn daily, auto-compounding interest. The lowest-stress way to put your idle deposit to work for 4% yield.
• An AI assistant in plain language. Ask questions, get market context, understand what you're looking at — built into the platform, no jargon required.
• An app that doesn't overwhelm. Faster, cleaner, and consistent — intuitive from the first tap, not after a tutorial.
• Safety you can check, not just trust. A MiCAR-licensed EU entity, Proof of Reserves you can verify yourself, and internationally certified security (SOC 2 Type II, ISO 27001:2022).

The BIS study adds an important perspective to the way Bitcoin on-chain transfer value is measured and interpreted. While every confirmed Bitcoin transaction is transparently recorded on the blockchain, converting those technical records into a measure of genuine economic activity requires assumptions about UTXOs, change outputs, self-transfers and wallet ownership. Depending on those assumptions, estimates of Bitcoin transfer value can differ by as much as sixfold.

For investors, researchers and market analysts, the finding reinforces the importance of looking beyond a single headline metric. Bitcoin transaction data can provide valuable insights into network activity, but its usefulness depends on how the underlying numbers are calculated. Comparing methodologies and combining transfer value with other on-chain indicators can provide a clearer picture of how Bitcoin is actually being used and how value moves through the network.

1. Is Bitcoin on-chain transfer value the same as trading volume?

No. Bitcoin on-chain transfer value measures BTC moving through blockchain transactions, while trading volume measures the value of Bitcoin bought and sold on exchanges or other trading venues. Large amounts of BTC can move on-chain without being traded, while heavy exchange trading can occur without an equivalent amount of Bitcoin moving directly across the blockchain.

2. Can Bitcoin on-chain data show who owns each wallet?

Not reliably. Bitcoin addresses are publicly visible, but they are generally pseudonymous rather than directly linked to real-world identities. Analysts can sometimes group addresses using known exchange wallets, transaction behavior and clustering techniques, but those methods involve assumptions and do not always identify ownership with certainty.

3. Why do different Bitcoin data websites report different numbers?

Blockchain analytics platforms can use different definitions, filters and address-clustering models. One provider may exclude certain self-transfers or internal exchange movements, while another may include them. As a result, metrics such as Bitcoin transfer volume, active addresses and exchange flows can vary even when platforms analyze the same blockchain.

4. What is the difference between transaction count and transfer value?

Transaction count measures how many transactions were confirmed on the Bitcoin network, while transfer value estimates how much BTC or dollar-denominated value moved through those transactions. A single large transaction can contribute heavily to transfer value without significantly changing transaction count, so the two metrics measure different aspects of network activity.

5. Can exchanges create large Bitcoin on-chain movements without customer trading?

Yes. Exchanges and custodians regularly move BTC between hot wallets, cold storage and operational addresses for security, liquidity and treasury management. These transactions can appear as large on-chain movements even when they are unrelated to new customer purchases or sales.

6. Does a large Bitcoin transfer always indicate buying or selling pressure?

No. A large Bitcoin transaction may represent an exchange transfer, institutional custody movement, wallet consolidation or a transfer between addresses controlled by the same entity. Analysts usually need additional information, including destination labels and exchange-flow data, before connecting a large transaction to potential buying or selling pressure.

7. What are address-clustering heuristics in Bitcoin analysis?

Address-clustering heuristics are analytical rules used to estimate whether multiple Bitcoin addresses may belong to the same owner. Researchers can examine shared transaction inputs, spending patterns and known wallet behavior to build clusters. These techniques can improve on-chain analysis, but they are probabilistic and can sometimes classify addresses incorrectly.

8. Can Bitcoin transaction fees affect transfer-value calculations?

Yes, although their impact is different from that of change outputs and self-transfers. Bitcoin transaction fees are paid to miners and are not returned as a normal output to the sender or recipient. Accurate analysis therefore needs to distinguish between transaction inputs, outputs and miner fees when estimating economic transfer value.

Disclaimer: This article is for informational purposes only and does not constitute financial, investment or trading advice. Crypto assets are volatile, and readers should conduct their own research before making financial decisions.