Don't want to trade it yourself?
Our desk runs DEX portfolios on profit share.
A transfer on the Bitcoin network currently costs around ten cents. Consolidating twenty separate amounts inside a wallet costs less than one euro. Both figures can be calculated directly from the current fee rate and the size of a transaction, and that is precisely why now is the moment to look into your wallet. This analysis was carried out by cryptoticker.io on September 9, 2026.
The starting position: the median fee rate on the Bitcoin network has sat almost continuously at one satoshi per virtual byte for three months, the technical minimum most nodes will relay at all. We evaluated 1,105 data points for this from the period between June 9 and September 9, 2026. At no single point did the rate exceed seven satoshis per virtual byte.
What the network fee in sat/vB actually is
The network fee of a Bitcoin transaction does not depend on the amount you send but on the space your transaction takes up in a block. That space is measured in virtual bytes, abbrever bids more per virtual byte gets in sooner
The fee rate is the price you bid per virtual byte, stated in satoshis per virtual byte, sat/vB for short. A satoshi is one hundred-millionth of a bitcoin. At a price of $78,495 or 67,496 euros per bitcoin, measured on September 9, 2026 at 18:51 UTC through the public price interface of CoinGecko, a thousand satoshis are worth around 67 cents.
From that follows a calculation with two factors: fee equals size times rate. An ordinary transfer with one input address and two outputs, so recipient plus change, takes up about 141 virtual bytes in the SegWit format common today. At one sat/vB it costs 141 satoshis and therefore around 0.10 euros. At two sat/vB it is 282 satoshis or about 0.19 euros. The amount sent plays no part in this: sending ten euros costs exactly as much as sending ten thousand.
Our measurement: 1,105 data points from three months of Bitcoin blocks
For this article we retrieved the fee rate ourselves rather than taking it from other outlets. The basis is the public interface of the block explorer mempool.space, endpoint /api/v1/mining/blocks/fee-rates/3m, retrieved on September 9, 2026 at 18:53 UTC with HTTP 200. The endpoint returns, for every data point, the distribution of fee rates of the transactions that were actually confirmed in blocks.
The scope: 1,105 data points across block heights 953,003 to 966,255, that is from June 9, 2026 at 19:20 UTC to September 9, 2026 at 18:22 UTC. In each case we evaluated the median of the fee rates within a point, because that value is more robust than the average: a single transaction bidding a hundred times the going rate out of urgency distorts an average, while it barely touches the median.
The result across the whole period: the median sits at one sat/vB, the mean at 1.36, and the range runs from zero to seven. The value zero stands for points in which half the confirmed transactions came in below a full satoshi per virtual byte, for instance because a miner included its own or bundled transactions. Nothing there was free of charge.
Why we look at the median and not the maximum
What counts for your decision is what a normal transaction cost, and not what the most impatient market participant was willing to pay. In our window the maximum reached three-digit levels at individual points, but each time it concerned only a handful of transactions. The median, by contrast, describes the price paid by the middle of the market, and that is the value you can orient yourself by.
90 percent of the data points sat at two sat/vB or below
The distribution across the 1,105 points comes out more sharply than the plain averages suggest. At 648 points, and therefore 58.6 percent, the median stood at exactly one sat/vB. At 239 points or 21.6 percent it was two, and at 112 points or 10.1 percent zero. Together that makes 999 of 1,105 points, or 90.4 percent, at no more than two sat/vB.
Above that the air thins out quickly: three sat/vB at 62 points (5.6 percent), four at 32 points (2.9 percent), five at seven, six at three and seven at two points. Not a single data point reached ten sat/vB. The more recent past shows no different picture either: over the last 30 days (361 points) the median was one and the maximum six, and over the last seven days (85 points) one and four.
There is one notable pattern, and it can be put to practical use. Sorted by time of day, the median sits at one sat/vB in 21 of 24 hours. Only in the hours around 15:00, 17:00, 18:00, 19:00 and 20:00 UTC does it rise to two. That is the window in which US trading hours and the European end of the working day overlap. By weekday, on the other hand, no difference could be measured: Monday to Sunday all sit at one sat/vB. So anyone with flexibility sends in the morning rather than in the early evening, and the difference in euros still amounts to only a few cents.
Hold your bitcoin safely yourself
What a UTXO is and why many small amounts get expensive later
Your wallet shows you a balance. On the blockchain that balance does not exist. What sits there are individual, mutually independent amounts, each tied to an address. Every one of these amounts is called a UTXO, short for unspent transaction output. What your wallet displays as your holding is the sum of your UTXOs.
This is the point at which the fee calculation becomes personal. Every UTXO you spend in a transaction becomes an input and takes up space. A SegWit input occupies around 68 virtual bytes. Anyone who has pulled a small savings-plan amount to their own wallet every week for two years has a good hundred such amounts sitting there. If many of them are to be spent at once one day, the sheer number of inputs alone costs a multiple of a normal transfer.
A consolidation is the answer to that: you send many small UTXOs to yourself in a single transaction and get one large one back. Throughout, the amounts stay in your own custody, for instance on a hardware wallet. What changes is solely the form in which your holding sits on the blockchain.
How to calculate the fee for your consolidation yourself
The formula is short enough to keep in your head. For transactions using the SegWit addresses common today it holds approximately that:
size in vB = 11 + (number of inputs × 68.5) + (number of outputs × 31)
An example with twenty inputs and one output gives around 1,412 virtual bytes. At one sat/vB this consolidation costs 1,412 satoshis and therefore about 0.95 euros. Five inputs to one output give 384 vB, so 384 satoshis or around 0.26 euros.
The comparison with a more expensive environment is where it gets interesting. The highest median in our three-month window was seven sat/vB. The same consolidation with twenty inputs would have cost 9,884 satoshis or around 6.67 euros there, so a good seven times as much. Should the rate climb to 50 sat/vB at some point, a value our measurement window does not contain and which we therefore state expressly as an arithmetic example and not as a forecast, it would be 70,600 satoshis or around 47.65 euros.
The real saving, though, only arises later, with the transaction in which you eventually spend that holding. If you then spend twenty individual UTXOs instead of one, you carry 19 additional inputs along. At seven sat/vB those cost 9,110 satoshis or around 6.15 euros, and in the 50-rate example 65,075 satoshis or about 43.92 euros. It is precisely this calculation that shifts in your favour when you place the consolidation in a phase of low rates.
Full mempool, low fee: why that is no contradiction
A glance at the mempool, the queue of transactions not yet confirmed, might be confusing at first. On September 9, 2026 at 18:52 UTC it held 81,263 unconfirmed transactions totalling 41.5 million virtual bytes. Arithmetically that corresponds to more than 40 full blocks, so roughly seven hours of work for the network.
What matters, however, is what that queue is bidding. Of the 41.5 million virtual bytes, only 0.92 million paid at least one sat/vB, and merely 0.42 million at least two. Everything else sits below that and waits for a miner with space to spare. For you that means the competition for the next block is smaller than one full block, and a transaction at two sat/vB stands practically at the front.
The fee recommendation from the same for the next confirmation, and one sat/vB each for half an hour, one hour and the economy setting. Anyone sending at one sat/vB should bring patience, though: with around 0.9 blocks of competition at the same level it can take several blocks, and a sudden jump in demand would push your own transaction back
Cheap does not automatically mean sensible. There are three reasons that can argue against it, and you should know them before you build a transaction.
Privacy. When you merge several UTXOs in one transaction, you make it publicly visible that these amounts belong to the same person. For anyone analysing the blockchain a link arises that was not documented before. Anyone deliberately keeping amounts from different
Tax treatment. Under the current administrative view, a transfer between two wallets of your own is not a disposal, and the holding period continues to run. Even so, you have to document the process cleanly so that your tax software assigns the amounts correctly later on. For larger holdings this does not replace tax advice.
The effort is out of proportion. Anyone with three UTXOs saves cents with a consolidation and risks an operating error in exchange. The benefit grows with the number of amounts, and below around ten inputs the exercise is rarely worthwhile.
What applies when your coins sit on an exchange
If your holding sits with an exchange, you have no UTXOs of your own at all, because custody lies with the provider. There you pay a fixed withdrawal fee when moving out, one that is only loosely connected to the actual network fee. We measured that on August 17, 2026 across 475 assets: for Bitcoin the exchanges charged eight to sixteen times what the transaction would have cost in the same minute. At a rate of one sat/vB, that mark-up weighs even more heavily than usual. A look at your provider’s fee page therefore belongs before every withdrawal, and anyone using several crypto exchanges will find marked differences there.
Setting the fee rate in your wallet: what to check before sending
Almost every wallet allows the fee rate to be set by hand, often behind a button for advanced settings. Four points are decisive here.
First: check the current rate at an independentculate with a safety mark-up that, at a rate of one sat/vB, amounts to a multiple of what is needed. The block explorer mempool.space displays the current recommendation without a login
Second: pay attention to whether your wallet states the fee as a rate in sat/vB or as an absolute amount. Only the rate is comparable. The technical basis for this, meaning how the weight and size of a transaction are calculated, is documented in the Bitcoin wiki on transaction fees.
Third: look for whether your wallet offers the replace-by-fee function. It lets you replace a transaction set too low with one carrying a higher fee later, rather than leaving it hanging indefinitely. Anyone sending at one sat/vB should have that option enabled.
Fourth: check the destination address of a consolidation especially carefully. You are sending to yourself, so a typo will not be caught by an indignant recipient. Anyone working with a withdrawal whitelist at their exchange already knows the principle.
Limits of this analysis: what the data does not show
The measurement rests on a single dataombine several blocks, so our 1,105 points cover 13,253 blocks without resolving them individually. We did not cross-check a second independent
What also cannot be derived is what individual users actually paid: the values describe confirmed transactions across the network as a whole, not German users in particular. The size estimates for transactions are likewise approximations for the SegWit format common today; anyone still using older address formats ends up with larger transactions and therefore higher fees. And finally, every one of these figures is a snapshot from September 9, 2026. A phase of high demand can multiply the rate within a few hours.
Checking the Bitcoin network fee: what to take away
- Look at how many UTXOs your wallet holds. Most wallets show this in the coin selection or under the advanced settings. From around ten individual amounts it is worth doing the consolidation arithmetic, and at a rate of one sat/vB the exercise costs you less than one euro even with twenty inputs. If you are thinking about moving into self-custody along the way, our comparison of hardware wallets helps with the choice.
- Check the current fee rate before you send. Compare your wallet’s suggestion against a block explorer before you accept it. And if there is no hurry, place a transaction in the morning rather than the early evening: in the hours between 15:00 and 20:00 UTC the median in our measurement was twice as high.
- Compare your exchange’s withdrawal fee with the real network fee. Fixed withdrawal fees stay where they are even when the network costs almost nothing. Anyone withdrawing regularly should lay the terms of the various crypto exchanges side by side and bundle the withdrawals instead of moving every small amount on its own.
(As of September 9, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)
Transparency note: This article was produced with the assistance of artificial intelligence and reviewed by our editorial team before publication. All figures and claims were checked against the primaryI
More from CryptoTicker
Source: cryptoticker.io